Polyvinyl chloride (PVC) pipes can be joined using mechanical joints or solvent
cement joints. Solvent cement joints are the most commonly used joints especially
for small diameter pipes because they provide a system with the least risk (Lu
et al., 2000). PVC cemented joints may fail due to cracks through the
adhesion between the socket and the pipe interface and through the damage in
the sockets or the pipes. When installing piping systems, some changes in the
direction may be necessary. So, the response of PVC pipe materials to longitudinal
bending is considered to acquire significant advantages especially in buried
and suspended applications. Longitudinal bending may be done deliberately during
installation to make changes in the alignment to avoid obstructions or it may
occur in response to unplanned conditions such as changes in soil properties
(Anonymous, 1991). PVC pipes with and without solvent cement joints were tested
in four points bending using variable internal pressure (Scavuzzo et al.,
1999, 1998). They found that the joined pipe specimens with no internal pressure
were weaker in fatigue bending than the pressurized pipes. Forte et al.
(1991) investigated the effect of bending radius and temperature on the service
life of Polyethylene (PE) pipe found that as the bending radius increased from
20 D (D = pipe outer diameter) to 40 D, the pipe life increased 13 times while,
the service life decreased by 80% due to a temperature rise of 10° C. Similarly,
Rahman and Mandal (1995) studied the effect of temperature and the internal
pressure sustaining capacity for PVC pipes and noted that there was a remarkable
effect of temperature on the mean failure pressure characteristics of PVC pipes.
However, with the increase of temperature, the mean failure pressure decreased
remarkably. Shi et al. (2002) investigated the effect of temperature
on the mechanical properties of under fill material and observed that the effect
of the temperature on the mechanical properties is much stronger at temperatures
around Tg than the temperatures below and above Tg. Chaoying
et al. (2004) studied the effect of temperature on the dynamic mechanical
properties of PVC with and without nano-CaCO3. They found that a
sudden decrease in the dynamic mechanical properties of PVC occurred at the
glass transition temperature. Wan et al. (2003) investigated the mechanical
properties of PVC-clay Nan composites at different temperatures. They found
that the mechanical properties decreased dramatically at glass transition temperature.
A dynamic mechanical analysis of glycol modified polyethylene terephthalate
(PETG) and poly methyl-methacrylate (PMMA) showed that a sudden change in the
mechanical properties of the materials under bending occurred at glass transition
temperature of each material (Jae and Cha, 2001). Aarkireyeva and Hashmi (2002)
examined the effect of temperature on the fracture parameter of uPVC film. They
found that the essential work of fracture is not affected by a temperature rise
up to 60° C which is the glass transition temperature of the used material.
The joints are points of weakness in the piping systems, so it is important to study the failure modes of joints when subjected to bending at different temperatures. The main objective of this research was to study the effect of temperature on the maximum bending force and the stiffness of PVC and CPVC cemented sockets to determine the best materials for installing piping system.
MATERIALS AND METHODS
The test specimens were prepared from PVC and CPVC pipes and the sockets manufactured in Saudi Arabia. The pipes were manufactured by extrusion while the sockets were made by injection molding. The PVC and CPVC pipes were manufactured according to the ASTM (American Society for Testing and Material) standards D 1785 and F 441, respectively. The outer diameter of pipes was 26.7 mm and the wall thickness was 4.3 mm. The sockets have outer diameters of 37 mm and a length of 56 mm. The test specimens for the study were designed as simply supported beams subjected to three points bending (Jae and Cha, 2001; Moosa and Mills, 1998). The distance between the beam supports was 420 mm. The bending force was applied at the middle of the beam by a hydraulic testing machine (Al-Naeem and AL-Hashem, 2005). The test specimens were fixed to the moving head of the testing machine while the loading rod and the load cell were attached to the fixed head. The bending force was monitored by using a load indicator connected to the load cell.
Three types of test specimens were tested. The first type of test specimens were pipe segments of 50 cm length. This test was conducted to quantify the effect of the socket joints on the stiffness. The second type of specimens were tested to quantify the contribution of the glue in the joint stiffness. In this test, the specimens were prepared by fitting the pipe segments into the sockets without glue (uncemented joints). The third type of test specimens were used as the main test in which specimens were prepared by using the solvent cement corresponding to each material (PVC and CPVC) as a glue material (cemented joints). In order to make sure that the joint glue reached its steady state, the specimens were prepared and leftover for 15 days before testing. The first and second types of specimens were tested at room temperature only. The third type of specimens were tested at six different temperatures namely 23, 50, 60, 70, 80 and 90°
To obtain the testing temperature of 50°
C and above, a homogenous temperature through the joint wall was obtained by immersing the test specimens in an automatic digital water bath for 24 h at the desired test temperature. Later on the specimens were taken out from the water bath and insulated to keep their temperature constant. Then, the test specimens were fixed immediately to the testing machine and the insulation material was removed. To keep the specimen temperature constant during testing, it was subjected to a jet of hot air from an electric heater. The distance between the heater and the specimens was adjusted to obtain the desired test temperature around the test specimen. The specimens were loaded up to a deflection of 55 mm (Al-Naeem and Al-Hashem, 2005). The speed of the moving head of the testing machine was adjusted at 2.5 cm min1 (Ollick and Al-Amri, 2000). Each test was replicated three times. The deflection of the test specimens was measured by fixing a ruler having 0.5 mm divisions to the moving head of the testing machine. The motion of the ruler related to a fixed line and the load indicator readings were recorded using a video camera. The video tape was played back using a special video and the play back speed was adjusted to be 1/16 of the recording speed. To increase the deflection measuring accuracy, the picture of the load reading and the ruler divisions was magnified 50 times by connecting the video to a data show. The output picture of the data show was displayed on 2x
2 m screen. The readings of the load indicator and the corresponding deflection were tabulated for each 1 mm deflection. The data representing the relationship between the bending force and the deflection were drawn to determine the bending energy. The bending energy is represented by the area under the load deflection curve; therefore, it was obtained by integrating the fitting equations. A QBASIC program was written to perform these integrations using 0 and 55 mm as the lower and upper integration limits, respectively.
The local stiffness (KL) of the test specimen at a certain point is defined as the rate
of change of the bending force (F) with respect to the deflection (Δ) at this point (KL
= δF/δΔ). In this study, the local stiffness was taken as the slope of the fitting line which fits the relationship between the load and the corresponding deflection over a deflection intervals of 2 mm (i.e., 0-2, 2-4, 4-6,..........
RESULTS AND DISCUSSION
Effect of temperature on the maximum bending force: Figure
1 and 2 shows the relationship between the bending force
and the corresponding deflection of pipes, cemented joints and uncemented joints
at room temperature for PVC and CPVC materials, respectively. The integration
results are shown in Table 1.
The bending energy of cemented PVC and CPVC joints (PVC and CPVC) was 31 and
53.3%, respectively more than the uncemented joints. i.e., the effect of CPVC
cement on the joint toughness was 1.72 times more than that of PVC cement (Table
1). Also, the bending energy of PVC pipes was 9.9% more than the CPVC pipes.
|| Relationship between the bending force and the corresponding
deflection of PVC cemented joint, uncemented joint and pipes
|| Relationship between the bending force and the corresponding
deflection of CPVC cemented joint, uncemented joint and pipes
In the case of un-cemented joints, the bending energy for PVC joint was 13.6%
more than that of CPVC joints. Similarly, for cemented joints, the bending energy
for CPVC was 2.9% more than that of PVC (The effect of CPVC cement inverted
the ranking of the other two conditions). These phenomena occurred because the
PVC pipe segments were observed to pull out the socket during loading. On the
other hand the pull out phenomena did not exist in CPVC joints. However, a bending
in the pipe segments was observed at the socket edge
The relationship between the mean bending force and the corresponding mean
deflection for PVC and CPVC joints at different temperatures (23, 50, 60, 70,
80 and 90° C) is shown in Fig. 3, 4,
respectively. It was also observed that the bending force and consequently the
bending moment capacity of PVC joints decreased with respect to the temperature
at higher rates than the CPVC joints. A sudden drop in the bending force capacity
of PVC joints was noted with the change in temperature from 60 to 70° C
(Fig. 3). This sudden drop exists because the glass transition
temperature of PVC material was within this temperature range [Tg = 60°
C (Shi et al. 2002), Tg = 62.5° C (Chaoying et al.,
2004), Tg = 72° C (Aarkireyeva and Hashmi, 2002)].
|| Bending energy (joules) for pipes and joints (uncemmnted
and cemented) of PVC and cemented) of PVC and CPVC at room temperature (23°C)
|| Relationship between the bending force and the corresponding.
deflection of PVC joints at different temperatures
|| Relationship between the bending force and the corresponding
deflection of CPVC joints at different temperatures
Since CPVC pipes are
used for higher temperature applications, it is expected that the effect of
temperature on the loading capacity must be lower than the PVC pipe material.
However, there was no sudden drop in the bending force capacity of CPVC joints
due to increase in temperature (Fig. 4).
Figure 5f describe comparison between the PVC and CPVC joints
in terms of force-deflection relationship. At room temperature (23° C),
the relationship for the two materials coincided up to a deflection of 25 mm.
In the case of deflection values more than 25 mm, the force-deflection relation
of PVC joints starts to deviate to lower values than those for the CPVC joints.
This behavior could be attributed to two reasons (1) due to the fact that the
plastic deformation of PVC is higher than the CPVC, i.e., the effective deflection
of CPVC is higher than the PVC material (effective deflection = total deflection-plastic
deformation). (2) it is related to the adhesion materials because the damage
in the adhesion between the pipe and the socket for PVC joints was observed
at lower deflection values than the CPVC joints. For the testing temperatures
of 50 and 60° C (Fig. 5 b,c)
and at the same deflection value, the bending force of PVC joints was lower
than the CPVC joints over the entire deflection range. It was also noticed that
the difference in the mean bending force between PVC and CPVC joints is approximately
the same for the two testing temperatures. With respect to testing temperatures
of 70, 80 and 90° C (Fig. 5d-f, respectively),
the difference between the mean bending force of the two types of joints increased
particularly at 90° C. Results further reveal that in the case of PVC joints
at the deflection values between 35 and 40 mm, the mean bending force decreased
with increase in the deflection values. These phenomena occurred because the
pipes were observed to be pulled out from the socket after these deflection
A relationship was developed between the mean maximum bending force and the
test temperatures for PVC and CPVC joints (Fig. 6). The data
of uncemented joints at room temperature was also included. The sudden decrease
in the maximum bending force of PVC joints between 60 and 70° C temperatures
occurred because the glass transition temperature of the PVC material is within
this temperature range (Shi et al. 2002; Chaoying et al., 2004;
Aarkireyeva and Hashmi, 2002). At room temperature, the mean maximum bending
force of the cemented joint of PVC and CPVC materials increased by 16 and 36%,
respectively than the un-cemented joints. It was further noticed that over the
entire temperature range, the mean maximum bending force of PVC and CPVC materials
decreased by 84 and 42%, respectively. However, in the case of PVC, 35% of this
percentage occurred due to the increase in temperature from 60 to 70° C.
Effect of temperature on the stiffness of joints: The results reveal that for the PVC pipe joints at higher temperatures and at a deflection value about 40 mm, the load decreased with the increase in deflection value (i.e., a negative stiffness). Therefore, the analysis of the two materials was limited to the deflection values ranging between 0 and 40 mm.
Figure 7 and 8 show A relationship was
developed between the deflection and the local stiffness of PVC and CPVC pipe
joints (Fig. 7, 8). It was found from the
data in two figures that the local stiffness was approximately constant for
the two materials (PVC and CPVC pipes) over the first stage of deflection at
different test temperatures, At the next stage of deflection, the local stiffness
started to decrease with an increase in the deflection value. The regions of
constant local stiffness seemed to extend over the deflection values equal to
that which correspond to the elastic limit (Ollick and Al-Amri, 2000), while
the regions of steeper slope correspond to the elastic-plastic regions. The
yield stress decreased with an increase in temperature, so the elastic region
interval decreased thus resulting in the downturn movement of point to the left
as shown in Fig. 7 and 8.
Figure 9 describe the relationship between the average values
of local stiffness in the elastic region and the test temperature both for the
PVC and CPVC joints. The data of uncemented joints and pipes at room temperature
was also considered. The slope of the fitting lines represents the rate of decrease
of stiffness with respect to temperature. Table 2 shows the
average decreasing rates of stiffness both for the PVC and CPVC cemented joints
at different temperatures.
|| Comparison between the bending force and the corresponding
deflection of PVC and CPVC joints at different temperatures (a) 23, (b)
50, (c) 60, (d) 70 (e) 80 and (f) 90° C
|| Relationship between the maximum bending force and test temperatures
of PVC and CPVC joints
|| Relationship between the local stiffness and deflection of
|| Relation between the local stiffness and deflection of CPVC
||Average decreasing rate of stiffness of PVC and CPVC joints
per a° C (N mm- 1 per ° C)
|| Relationship between the average stiffness and test temperature
in the elastic loading region for cemented and un-cemented joints and pipes
However, at any given temperature range, the decreasing rates of mean stiffness
of PVC is higher than the CPVC. The decreasing rate of PVC joints changes in
non-uniform manner, whereas, the decreasing rate of CPVC increases gradually
with the increase in temperature. At the temperature range of 23-50° C,
the decreasing rate of mean stiffness of PVC joints was 2 times higher that
the CPVC joints. Which means that CPVC joints were not sensitive to the corresponding
change in temperature at the lower temperature levels. However, at a temperature
range between 60-70° C, the ratio between the decreasing rates of mean stiffness
of PVC and CPVC joints increased again to 2.1 which could be related to the
effect of the glass transition temperature of PVC material.
At room temperature, the local stiffness of the cemented PVC or CPVC joints
was 50 and 74% higher, respectively than the un-cemented PVC or CPVC joints
On relative basis, the percentage decrease in the stiffness at different temperatures
relative to the stiffness at room temperature (23° C) is presented in Table
3 and Fig. 9. The values of PVC joints show that the joint
stiffness decreased approximately 1% per ° C with the change in temperature
from 23 to 80° C. whereas, the stiffness decreased up to 2.1% per °C with the change in temperature from 80 to 90° C.
|| Comparison between the local stiffness and the corresponding
deflection of PVC and CPVC joints at different temperatures (a) 23, (b)
50, (c) 60, (d) 70, (e) 80 and (f) 90° C
||Mean percent decrease in the stiffness of PVC and CPVC joints
relative to the stiffness at room temperature
However for the CPVC
joints, the joint stiffness decreased approximately 0.5% per ° C with the
change in temperature from 23-80° C, whereas, the stiffness decreased by
1.44% per ° C with the change in temperature from 80 to 90° C.
Figure 9 shows a comparison between the stiffness of pipes,
cemented and uncemented joints of PVC and CPVC materials at room temperature.
For pipes and uncemented joints, the difference between the average stiffness
of PVC in elastic region and that of CPVC was 4.5 N mm- 1. Whereas,
for the cemented joints of PVC and CPVC joints, this difference was only 0.6
N mm- 1. The constant difference in case of pipes and uncemented
joints existed because the stiffness depends on the properties of the materials.
While in the case of cemented joints, the stiffness depends jointly on the properties
of both the materials and the solvent cement properties. This suggests that
the contribution of CPVC cement in the stiffness of joints was higher than the
contribution of PVC cement.
Figure (10a-f) describe a comparison
between the local stiffness of PVC and CPVC cemented joints and the corresponding
deflection at different temperatures. As mentioned earlier, the stiffness-deflection
relationship has two stages. The first stage has constant stiffness and corresponded
to the elastic-deflection region, while the second stage has steeper slope and
corresponded to the elastic-plastic deflection region (Ollick and Al-Amri, 2000).
At room temperature the two joint materials approximately coincide over the
elastic region while in elastic-plastic region, the stiffness of PVC joints
deviates and has low values. As the temperature increases, three phenomena take
place. The first one is the movement of the stiffness-deflection relation of
the two materials to lower values. The second is the difference between the
stiffness of the two materials which increased with increasing temperature.
The third is the point of down turn which moves to the left with an increase
in temperature thus indicating a shorter elastic region. Since the elastic region
depends on the yield stress of the material and consequently the yield stress
decreases with increasing temperature and this will ultimately reduce the elastic
The decreasing rate of the mean maximum bending force and the mean stiffness
of PVC joints was higher than the CPVC joints at different temperatures. With
the increase in temperature from the room temperature to 90° C, the mean
maximum bending forces of PVC and CPVC joints reduced to 16 and 58%, respectively.
The mean stiffness of PVC and CPVC joints reduced to 20 and 52.4%, respectively
as compared to the mean stiffness at room temperature. However, at any given
temperature, the contribution of CPVC cement in the stiffness of joints was
higher than that obtained for the PVC cement. Based on the study results, CPVC
cemented joints are recommended for used in most cases rather than the PVC cement
joints, especially for the buried and suspended applications to provide long
The authors thank the Saudi Basic Industrial Company (SABIC), Saudi Arabia, for financial support to carry out this research.