April
12,
2005
NOTE
SUBJECT:
Evidence
of
Condensed
Moisture
for
HWCs
w/
Wet
Scrubbers
FROM:
Bob
Holloway
EPA/
OSWER/
OSW/
HWMMD/
WTB
TO:
The
Docket
 
OAR­
2004­
0022
On
April
7,
2005,
I
called
Mel
Keener,
Director
for
the
Coalition
for
Responsible
Waste
Incineration
(
CRWI),
to
ask
if
he
could
document
by
engineering
information
or
data
that
there
is
condensed
moisture
in
stack
gas
for
HWCs
controlled
w/
wet
scrubbers.
This
request
is
in
response
to
comments
from
CRWI
and
Ely
Lilly
and
Company
(
Lilly)
at
proposal
that
water
droplets
are
present
at
levels
that
could
cause
a
low
bias
in
HCl
and
Cl2
emissions
measured
w/
M26.

Mike
Foster,
Lilly,
responded
with
the
attachment.
ATTACHMENT
Message
to
Bob
Holloway,
EPA
Via
e­
mail
4/
8/
05
Mel
passed
along
your
request
for
examples
of
situations
where
there
are
water
droplets
in
the
stacks.

In
my
experience
in
performing
stack
tests,
this
situation
is
most
prevalent
on
incinerators
with
simple
scrubbing
systems
consisting
of
an
adiabatic
quench
followed
by
a
venturistyle
scrubber.
Adiabatic
quench
temperatures
are
typically
180
+/­
20
oF.
At
these
temperatures,
the
dry
gas
carries
large
amounts
of
moisture
as
evidenced
by
the
steepness
of
the
saturation
line
on
a
high
temperature
psychrometric
chart.
This
steepness
also
indicates
that
even
slight
amounts
of
cooling
after
the
scrubber
can
result
in
the
formation
of
significant
free
water
in
the
gas
stream.
In
a
real
stack
with
laminar
flow
(
which
is
what
is
desired
for
stack
sampling),
the
cooling
first
takes
place
along
the
stack
walls
with
the
rate
of
heat
transfer
being
a
function
of
the
stack
gas
temperature,
the
stack
material,
and
the
ambient
temperature.
Thus,
it
is
easy
to
see
how
a
saturated
gas
exiting
a
scrubber
can
develop
water
droplets
on
the
way
up
the
stack
to
the
sampling
probe.
Often
times
the
water
will
run
down
the
stack
wall
and
then
drip
across
the
stack
intake
opening
where
the
high­
velocity
stack
gas
will
entrain
the
droplets.

During
a
stack
test,
the
most
obvious
indication
of
the
presence
of
such
droplets
is
that
the
impingers
in
the
sampling
train
will
collect
excessive
amounts
of
water.
Experienced
stack
samplers
will,
in
such
situations,
calculate
the
stack
gas
moisture
percentage
based
on
the
impinger
catch,
and
then,
as
a
check,
perform
a
psychrometric
calculation
to
determine
the
saturation
moisture
content
(
i.
e.
the
moisture
content
at
100
percent
relative
humidity).
The
lower
moisture
content
calculated
is
then
used
in
subsequent
calculations
to
determine
emission
levels.
In
other
words,
if
the
moisture
content
determined
from
the
impinger
catch
exceeds
the
saturation
level,
then
water
droplets
were
sucked
into
the
sampling
probe
during
the
test
run.

Lilly
had
an
example
of
this
phenomenon
on
our
Trane
incinerator
at
Clinton
during
a
trial
burn
in
1986.
(
This
source
is
included
in
the
HWC
MACT
database).
The
data
are
as
follows:
Run
#
Stack
Temperature
o
F
Measured
Moisture
(%)
Saturation
Moisture
(%)
Water
Droplets?
6­
1
170
43
41.3
Yes
6­
2
160
33.9
32.7
Yes
6­
3
161
34.5
33.4
Yes
Making
a
cursory
inspection
of
the
sources
in
the
database
that
CRWI
perceives
are
the
HCl/
Cl2
top
performers,
I
note
that
sources
349,
478,
and
706
are
operating
at
the
saturation
level
and
very
likely
have
water
droplets
in
the
stack.
The
data
are
as
follows:
Source/
Run#
Stack
Temperature
o
F
Reported
Moisture
(%)
Saturation
Moisture
Range
(%)*
Water
Droplets
Likely?

349C11/
1
181
53.4
50.4­
53.9
Yes
349C11/
2
181
54.1
50.4­
53.9
Yes
349C11/
3
180
53.2
49.3­
52.3
Yes
706C2/
1
183
54.3
52.7­
56.4
Yes
706C2/
2
184
51.8
53.9­
57.6
No
706C2/
3
185
56
55.1­
58.9
Yes
478C10/
1
184
55.1
53.9­
57.6
Yes
478C10/
2
185
54.8
55.1­
58.9
Yes
478C10/
3
183
55.2
52.7­
56.4
Yes
*
Range
is
due
to
saturation
being
a
function
of
barometric
pressure
and
static
pressure
which
are
not
available
in
the
database.
Barometric
pressure
range
of
29
to
31
inches
assumed
with
0.0
static
pressure.

I
hope
this
answers
your
question.
Please
contact
me
if
you
need
further
explanation.

Mike
Foster
Eli
Lilly
and
Company
317­
277­
1094
