Table 7. Roof cut C, test results.
About 5 in. of FGL was present below this. The lower 1/2 in. was frozen to the deck and
the rest broke apart into small pieces during sampling, so it was not retained for testing.
Core 4 sampled this material.
faced bituminous built-up membrane, perlite in-
with dry insulation before the membrane was
sulation (3 in. [7.6 cm]), fibrous glass insulation (5
patched. Table 8 presents findings of core 9 and
in. [12.7 cm]), and a concrete deck. The tempera-
Table 9 presents findings of cut D. The perlite is
ture of the freezer below this roof was 0F.
essentially dry and has retained most of its insu-
The Hypalon was not well installed and numer-
lating ability. Portions of the fibrous glass insula-
ous wrinkles are present as shown in Figure 18.
tion are wet, and that moisture has reduced its in-
There is one bare spot on the roof where ballast
sulating ability. The lack of well-defined thermal
does not protect the Hypalon. It is shown in Fig-
anomalies suggests that this level of wetness is
ure 19. The membrane is also wrinkled there. There
present over most of the roof.
were no bright or dark thermal anomalies on this
No flaws were evident on this roof. It would
roof. Core 9 and cut D were taken near the center
take a more detailed investigation to determine
of this roof where shown in Figure 2. Figure 20
how the fibrous glass insulation became wet with-
shows the amount of dirt present on the Hypalon
out wetting the thick layer of perlite insulation
membrane. A considerable amount of effort was
above. The fact that the perlite is dry suggests that
needed to clean the membrane so that it could be
the moisture in the fibrous glass was there before
cut and patched properly. Figure 21 shows the cut
the Hypalon membrane was installed over the
made when taking cut D. That hole was backfilled
original bituminous built-up membrane. It also
Figure 18. Raised edge of roof D5 showing wrinkled Hypalon.