materials through which the passage of water va-
Table 4. Statistical analysis of puncture test re-
por may be of importance. There are two basic
sults.
methods, the Desiccant Method and the Water
Standard
Coefficient
Manufacturer
Average
deviation
of variation
Method. These methods are provided for the mea-
surement of permeance, and two variations in-
First set of 20 tests
clude service conditions with one side wetted and
Soprema
157.7
13.6
8.6
service conditions with low humidity on one side
NEI
88.4
9.3
10.5
and high humidity on the other. Some suggested
Polyguard
71.6
10.6
14.8
Royston
50.3
4.0
7.9
standard test conditions from E96 are listed be-
Protecto Wrap
38.5
4.1
10.7
low:
W.R. Grace
36.5
6.7
18.3
Procedure A Desiccant Method at 73.4F
Column average
73.8
8.0
11.8
(23C)
Procedure B Water Method at 73.4F (23C)
Second set of 20 tests
Soprema
148.5
12.3
8.3
Procedure BW Inverted Water Method at
NEI
99.5
16.3
16.4
73.4F (23C)
Polyguard
53.8
7.0
13.0
Procedure C Desiccant Method at 90F
Royston
51.7
10.0
19.3
(32.2C)
Protecto Wrap
39.7
3.4
8.6
Procedure D Water Method at 90F (32.2C)
W.R. Grace
36.0
6.3
17.5
Procedure E Desiccant Method at 100F
Column average
71.5
9.2
13.8
(37.8C) and a 90% RH
Third set of 20 tests
Soprema
142.6
16.1
11.3
Procedures A through D are normally con-
NEI
90.9
10.4
11.4
ducted in a test chamber at 50% relative humidity.
Polyguard
70.0
12.5
17.8
Agreement should not be expected between results
Royston
42.1
4.4
10.5
obtained by different methods and that method
Protecto Wrap
37.5
7.1
19.0
should be selected which more nearly approaches
W.R. Grace
39.0
8.1
20.7
Column average
70.3
9.8
15.1
the conditions of use.
A "perm" is the inch-pound unit of measure-
Fourth set of 20 tests
ment for permeance. One "perm" is the mass rate
Soprema
149.6
14.0
9.4
of water vapor flow through one square foot of
NEI
92.9
12.0
12.9
material or construction of one grain (gr) per hour
Polyguard
65.1
10.0
15.4
Royston
48.0
6.1
12.7
induced by a vapor pressure gradient between two
Protecto Wrap
38.5
4.9
12.6
surfaces of one inch of mercury or in units that
W.R. Grace
37.2
7.0
18.9
equal that flow rate (JTI 1990):
Column average
71.9
9.0
13.7
1 perm = 1 gr/h ft2 in. Hg =
5.75 1011 kg/s m2 Pa.
Historically, a material or system with a per-
among the bottom three membranes, because the
meance of 1 perm or less qualifies as a vapor re-
magnitudes of their data scattering exceed the dif-
tarder. More recently, further classification of va-
ferences in their puncture resistance.
por retarders has been proposed. For example, the
Canadian General Standards Board (CGSB) has
WATER VAPOR PERMEANCE
specified Type I vapor retarders as ones with a
permeance of 0.25 perm or less, and Type II as re-
This study reviewed and evaluated the suitabil-
tarders with a permeance of 0.75 perm or less be-
ity of the ASTM E96-95 Standard Test Methods for
fore aging and 1 perm or less after aging (ASHRAE
Water Vapor Transmission of Materials to sheet mem-
1997).
branes used as waterproofing between concrete
Table 5 summarizes the membrane manufac-
bridge decks and asphalt overlays used as the traf-
turers' data for the six membranes tested. The data
ficking surface. A modified test method, which is
are sorted from lowest to highest permeance and
under consideration for inclusion in ASTM E96,
shows that these values range from 0.003 to 1 perm
was investigated and the results compared to a
and vary by a factor of 333. The membrane thick-
variation of one of the standard test methods.
ness varies from 60 to 170 mils. The manufactur-
The test methods in ASTM E96 cover the deter-
ers generally tested their membranes using E96
mination of water vapor transmission (WVT) of
15
Previous Page
