Water Penetration Test Results for Vertical Joint
A summary of the results from water penetration tests on the vertical joint is provided in Fig. 13. In this summary, the degree of water penetration is given in terms of water leakage (L/min) as a function of pressure difference across the test specimen (Pa) for 16 test conditions for which the crack length was 2, 4, 8, and 16 mm and the joint displacement (Д) varied from 0 to 2 mm (Д =10 % joint width).
Quantity and Pattern of Water Leakage—The quantity and pattern of water leakage is very complex because the parameters affecting leakage are interrelated. However, in general, greater rates of water leakage occurred through cracks in the sealed joint, given either higher water deposition rates on the specimen surface or higher pressure differentials across the specimen.
The pattern of water leakage and the effect of the backer rod on water entry across the joint are characterized in Fig. 14. Specifically, when the diameter of the backer rod (nominal size 20 mm) was greater than the joint width, water flowed downwards along the gap between sealant and backer rod which is illustrated as Type “A” water leakage in Fig. 14. Whereas, when the rod diameter was less than the joint width, i. e., when the joint was extended from its original width of 20 mm, water flowed from the opening across the gap at the
substrate-backer rod interface, towards the back of the joint; this pattern of water leakage is illustrated as Type "D" in Fig. 14. When the rate of water entry was less than or equal to the drainage capacity along the gap, water flowed down the gap as in Pattern "B," (Fig. 14), whereas, if the rate of entry exceeded the drainage capacity along the gap, excess water flowed up the gap as in Pattern "C" (Fig. 14).
FIG. 11—Crack opening of vertical sealed joint.
Discussion of Water Penetration Results for Vertical Joint
The discussion focuses on two principal aspects that affect water leakage across deficient vertical joints, specifically, the influence of joint displacement and crack length, and the effect of the backer rod. Each is dealt with in turn.
Influence of Joint Displacement and Crack Length—Given the presence of a crack, at no joint displacement (0 mm), water leakage nonetheless occurred. Additionally, and excluding results obtained with no joint displacement (0 mm), the longer the crack length and the greater the joint displacement, the greater the rate of water penetration. Hence, the crack length and joint displacement provided a multiplicative effect on water leakage rates. As well it can be observed that at the largest joint displacement, water readily penetrated the joint, even at low pressure conditions. From this, it follows that the greatest water penetration rate (0.05 L/min) occurred at the maximum test joint extension (2 mm), largest crack length (16 mm), and highest test pressure differential (2000 Pa).
Some additional observations from the results include the following:
• The greater the driving pressure across the joint, the greater the rate of water penetration.
• When the crack length is “small" (e. g., <2 mm), regardless of the rate of water deposition at the crack location, at the low pressures, there are few differences in the rate of water penetration across the joint.
• However, given the “small" crack lengths, at high pressure differences
across the joint, increases in the rate of water penetration are evident for corresponding increases in the rate of water deposition at the crack location.
• When the crack length is large (e. g., 16 mm), the higher the rate of water deposition at the crack, the greater the rate of water leakage.
• At larger joint displacements (i. e., 5 and 10 %) and for low water deposition rates at the crack (i. e., 1.6 L/(min-m2)), even at high pressure differences, the increase in the rate of water leakage with increase in pressure difference are not as significant as compared to leakage rates achieved at more substantial water deposition rates; hence occlusion of the opening with water is a factor that affects gross water entry rates.
Effect of Backer Rod—The closed-cell polyethylene backer rod acted as a gasket providing a secondary barrier to water entry at the joint. When leakage occurred at the crack, and given that the backer rod was compressed, water leakage was arrested beyond the position where the backer rod interfaces with the substrate; this typically occurred in conditions of no joint extension or when the extension was no greater than 0.5 mm. However, when the joint was extended (e. g., 1 and 2 mm) and the width of the backer rod was then smaller than the extended joint width, the rod no longer acted as a gasket. It is to be noted that the backer rod (nominal diameter 25 mm) once compressed in the 20 mm joint did not recover its size upon extension of the joint. Hence the degree of compression set of the rod affected its capability to seal the jointing
0 500 1000 1500 2000 0 500 1000 1500 2000 0 500 1000 1500 2000 0
Pa Pa Pa
FIG. 13—Water penetration test results for vertical joint. Variation in water leakage rates (L/min) in relation to pressure different across specimen (Pa) for 16 test conditions for joints having crack length deficiencies of 2, 4, 8, and 16 mm and joint displacements (Д) of 0 to 2 mm (Д =10 % joint width).
product. It appears critical to consider the diameter of backer rod and its degree of compression in a joint to help avoid water leakage should a fail-safe system be of interest.
There may also be consideration as to whether an open or closed-cell backer rod would greatly affect water leakage when the rod is sufficiently compressed as to arrest the flow of water; it is expected that an open-cell rod would deter water entry but perhaps not as effectively as a closed-cell backer rod. However, such a notion would have to be evaluated from results of further testing.
FIG. 14—Characterization of water leakage of vertical joint.
FIG. 16—View of underside of horizontal joint showing backer rod and location of crack (16 mm) opening along jointing product; arrow shows direction of water leakage across joint.