A clear filtrate is important because it indicates a
used in these tests had almost twice the initial sol-
good solids capture by the filter cloth and a low
ids content.
solids loading in the filtrate return flow to the
In addition to removing water in the form of fil-
plant. The clarity of the filtrate was measured with
trate, the vacuum filter must also accumulate and
a turbidimeter. We selected eleven cloths based on
retain a uniform layer of sludge on the filter cloth
their porosity values. Those with a low porosity
for subsequent freezing. Tests conducted with
were assumed to have a greater potential for pro-
cloth 2019 (see Fig. 11) indicate that the weight of
ducing a clearer filtrate.
the sludge layer increased as the filtration time
Each filter cloth was tested at a vacuum pres-
was increased from 0 to 4 minutes. Longer filtra-
sure of 100 mm of Hg, and filtration times of 2, 5,
tion times caused a decrease in the sludge layer.
and 10 minutes. As shown in Table 1, the clearest
The reason for this decrease is not clear. Perhaps
filtrate was produced by cloth 515, although cloths
some of the attached solids were released as the
201, 208, 210, 2015 and 2019 all had turbidities less
layer compressed due to prolonged filtration.
than 30 NTU, which is reasonably clear. As expect-
ed, the clearer filtrates were produced by the
Characteristics of sludge layer
cloths with lower porosities. The type of material
The thickness of the sludge layer on the filter
or the weave of the cloth did not seem to make a
cloth ranged from 1 to 7 mm. The total solids con-
difference.
tent in these layers ranged from 1.0% to 6.9%. The
main determining factor for layer thickness seems
Vacuum level
to be the solids content in the initial sludge. A high
To conserve energy, the vacuum level applied
solids content in the sludge translated into a layer
should only be high enough to form a sludge layer
of greater thickness and higher solids content.
on the filter cloth while significantly reducing the
After the sludge layer was frozen, it was remov-
volume of sludge to be frozen. To determine this
ed from the filter cloth as described earlier. The
optimum level, filter leaf tests were conducted
layer was mostly opaque except in the areas where
with cloth 515 at 100, 200, 300, 400, and 500 mm of
transparent ice crystals penetrated the sludge lay-
Hg. The total solids content in the alum sludge
er from top to bottom. As each sludge cake
was 1.14%, and the filtration time was set at 2.0
thawed, the clear meltwater quickly drained
minutes. The results of these tests indicate that in-
away. When thawing and drainage were com-
creasing the vacuum level did not significantly in-
plete, all that remained was a small deposit of
crease the volume of filtrate (see Fig. 9). This be-
granular solids. The average solids content of this
havior is indicative of a compressible sludge layer
deposit was 11.8%. If left to dry overnight, the sol-
where particles are drawn into a more compact ar-
ids content reached 70% or more.
rangement by the increased vacuum level. The
One of the most important operational con-
only benefit from a higher vacuum level appears
cerns when using vacuum filtration is the condi-
to be a more rapid rate of solids accumulation on
tion of the filter cloth after several cycles. Based on
the media surface.
visual observations, the filter cloth separated very
cleanly from the frozen sludge layer as long as the
Filtration time
layer was completely frozen (see Fig. 12). To evalu-
The filtration time is important because it deter-
mines the drum speed at a fixed submergence lev-
el. For practical purposes, it should be long
enough to coat the filter cloth with a thick layer of
sludge while continuously removing filtrate. Six
different filter cloths were tested (see Fig. 10) at
three filtration times: 2, 5, and 10 minutes. Five of
the filter cloths produced similar results in that the
volume of filtrate increased as the filtration time
increased. The only exception was cloth 2015,
which showed a decrease in filtrate volume after 5
minutes, presumably because of a more rapid sol-
ids buildup due to a higher initial solids content.
For the same reason we suspect that cloth 2019
Figure 12. Cloth 2019 after removal of sludge layer.
produced low filtrate volumes because the sludge
Note clean separation of sludge layer from cloth.
11
Previous Page
