Page 1
Carbon monoxide:
A mechanic’s approach
Application Note
It’s not always technical maid service. Sometimes HVAC professionals
are rewarded with a mystery and
given the opportunity to put our
technical detective skills to work.
How accurately we solve combustion
related mysteries is directly related
to our fundamental understanding
of the science and technologies in
effect, our choice of tools and test
equipment, our previous experiences, and even our imagination.
The answers lie beyond
the obvious
Carbon monoxide consumer
awareness is still on the
increase. More and more
consumers are installing CO
detectors in their homes and
workplaces. This is good. It
offers occupants a level of safety
they did not have previously,
and offers the expert (you) an
opportunity for in-depth site
analysis.
So how does it usually unfold?
An occupant calls the fire
department after a CO alarm.
The firemen may not see an
obvious source of the CO, so they
recommend having the furnace
(or boiler) checked by a heating contractor: the combustion
expert. It’s your job to find the
source of the CO under current
conditions, or to try to duplicate
variable conditions under which
CO could be produced. This is a
tall order, requiring a process of
elimination of possible malfunctions and a keen awareness
of variables, such as building
depressurization and gasses (air
and vent products), that could
behave unexpectedly.
While you’re driving to the
jobsite, turn off the radio and
think about the abundant possibilities. Think of the structure
in its entirety as a system.
Combustion appliances are only
a part of the integrated whole.
What else is competing for or
interfering with combustion air,
fuel supply, vent function? How
can contaminants such as chlorides, sulfides, VOC’s, dust, and
dirt affect component integrity
and operation?
Think about some of the
questionable creative engineering you’ve seen from end users
trying to save on energy costs.
Think about all of the different
fossil fuel appliances and products, wood fireplaces and stoves
that may be in play: some fixed,
some portable, some that drive
away. Think abo ut what could
be external to the structure
that could produce combustion
products and be drawn into the
occupied space: idling school
buses or trucks, standby generators in exercise mode, poorly
placed or improperly installed
vent terminations.
The list of possibilities is
seemingly endless, and more
often than not, it’s two or more
processes gone wrong. One
process may have been operating on the edge of acceptability
since day one, and a second
process changes enough to
throw the first process over the
edge. Every job is different and
valuable lessons can be learned
from each. Be observant and
allow your mind to explore the
possible system failure modes
and their interacting effects.
Exterior survey
When you arrive at the jobsite, observe the exterior of the
system (the building and its
immediate surroundings). Before
entering the building, turn on
your environmental air meter
(such as Fluke 975) and give it
time to warm up and zero outdoors. This will give you time
for a walk-around.
F r o m t h e F l u k e D i g i t a l L i b r a r y @ w w w . f l u k e . c o m / l i b r a r y
Page 2
Notice chimneys, vertical
vents, sidewall vents, air intakes
and exhausts, exhausts that
could become intakes (such as
drier vents that stick open from
lint build-up), building penetrations proximity to doors and
windows, and meter locations.
Look underneath decks and
porches, at the garage location,
inside corners for vents, and
notice shrubberies and trees,
prevailing wind direction and
possible effects, and foundation
type (slab, crawlspace, basement). Look in window wells
and crawlspace access wells,
and just generally notice things.
Later, as you focus on individual components such as the
furnace or boiler, water heater,
venting and ventilation, you
will return to the outdoors to
pay more attention to the details
and interactions of functions.
You might be surprised by how
a seemingly unrelated outdoor
feature can affect the indoor
function of a process.
Questions to ask
Once inside the structure, consider yourself as a combination of
police detective and crime scene investigator.
Here are some questions to ask:
What kind of CO alarm sounded?
•
When did the alarm sound?
•
Was it a standard CO detector
•
designed according to UL 2034 that
specifies that alarm must sound
within 1 to 4 hours at 70 ppm CO, or
do they have a low level alarm model
that sounds after 5 minutes when CO
reaches 15 ppm to 34 ppm?
Can you retrieve the maximum CO
•
level retained in the alarm memory?
What CO level did the firemen find?
•
What other comments did the fire-
•
men have besides suggesting to call
the heating contractor?
Did the alarm sound during early
•
morning hours when all were asleep
indicating possible ventilation and
venting problems?
Did any of the occupants exhibit any
•
physical symptoms?
Did it sound after everyone was
•
awake indicating a possible occupant created condition from space
heaters or automobile warm up in an
attached garage?
Did it sound on Sunday afternoon
•
when friends or relatives came for
dinner?
Did they burn the roast, burn the
•
biscuits, or burn the food in the pan
on a burner they forgot about?
Is there an exhaust fan in the kitchen
•
over the stove and is it used?
Was an outdoor cooking grill used
•
and at what location?
What about gasoline powered tools
•
and equipment?
Is smoking allowed indoors?
•
Did it sound shortly after everyone
•
retired for the night and was the
fireplace used on that night?
Check more than just CO
Before you leave the outdoors,
record the outdoor ambient CO
so you can compare it to indoor
levels. Outdoor levels could be
as low as 380 ppm CO
or less
2
in rural and seacoast locations, or above 500 ppm CO
congested urban locations. Then,
as you start your indoor survey,
record CO
humidity on your air meter.
LEED-EB IEQ2
ppm, CO ppm and
2
(1)
Credit 1 CO2
concentrations can range from
less than 1,000 ppm CO
2,000 ppm CO
depending on
2
2
per person ventilation requirements and Met (metabolic) rate
activity levels.
(2)
Elevated CO2
(from respiration) can be an
indicator of a poorly ventilated
structure, but unvented (spilling) combustion products will
elevate CO
and humidity levels
2
in the occupied space as well.
Increased CO
levels in kitchens
2
may be due to decomposing
organic material (garbage).
CO
and water (vapor) are
2
produced in the complete combustion of fossil fuels, and CO
2
in
2
to over
and water (vapor) are produced
from incomplete combustion. A
furnace, boiler or water heater
typically produce 80,000 ppm
to 30,000 ppm (8 % to 13 %)
CO
depending on fuel type,
2
and about a gallon of water per
100,000 Btu input. If a category
I vent is spilling combustion products, the first change
an occupant may notice is
increased condensation on windows due to elevated humidity
that hadn’t occurred previously.
The first rule of fire
Fire must be safely confined
and controlled while maintaining correct fuel supply, oxygen
supply, and ignition temperature.
When we build a fire indoors,
we must know that the combustion products will vent to the
outdoors. Appliances designed
to operate over extended periods
of time like a furnace, boiler or
water heater must be vented
directly to the outdoors. Products
with limited operating times
such as residential stoves, ovens
and space heaters are usually
expected to have products of
combustion in quantities low
enough to be vented by normal
building ventilation, which
may be relying entirely on the
natural infiltration of air through
structural leaks. As oxygen is
consumed and vent products
exit the building, fresh air from
outdoors must be continuously
supplied or disaster will follow.
Combustion analysis
Warm up and zero your com-
•
bustion analyzer outdoors.
Combustion products must be
•
sampled undiluted before any
draft hoods, barometric draft
dampers, or any dilution air
injection. (Some high efficiency direct vent boilers may
draw air from the air intake
into the combustion air blower
prior to the exhaust pipe.)
Sample the O2 and CO on
•
startup and monitor as the
process settles in to steady
state operation. CO may be
high on startup and reduce to
2 Fluke Corporation Carbon monoxide: A mechanic’s approach
Page 3
normal values as steady state
operation is reached. Allow
at least 15 minutes for steady
state operation to be achieved.
Think about what you are
•
measuring. To find the amount
of excess air (air that has not
been used in the combustion
process), we measure O
CO
. To discover the quality of
2
2
or
combustion, we measure CO.
Since before any of us started
•
in this trade, the national
standard for maximum CO in
the vent of vented products
has been 400 ppm CO air
(3)
free.
Air free means: if we
remove the excess air from
the sample, what would the
CO be? Or, what would the
CO reading be with stoichiometric air: Only enough air for
perfect combustion without
any excess air whatsoever. No
product may legally exceed
this value, but some older
equipment may operate on the
threshold of this value.
Target values for modern
•
natural gas, LP gas and #2
fuel oil equipment steady state
maximum CO sample levels
should not exceed 100 ppm
CO. Setup values for efficiently operating equipment
is considered to be between
10 ppm CO and 100 ppm
CO. These values are actual
sample levels not corrected to
“air free” levels. This is a good
rule of thumb and is recommended unless specifically
overridden by the equipment
manufacturer.
O2 and CO2 are excess air
•
measurements and vary
widely between products.
O
is measured by modern
2
electronic analyzers and is
basically a direct indicator
of excess air. Six percent O
2
equates to about 40 % excess
air whether the fuel is natural
gas, LP gas, or # 2 fuel oil.
For comparison, 6 % O
and
2
40 % excess air is equivalent
to (using typical ultimate CO
6 % to 9 % O2 11 ppm to 99 ppm
CO for Bunsen type (primary air
and gas mixed injection) burners
- 6.8 % to 8.6 % CO
- 7.8 % to 9.8 % CO
3 % to 6 % O
99 ppm CO for power or
specialty burners (oil or gas)
- 8.6 % to 10.3 % CO
- 9.8 % to 11.7 % CO
- 11.1 % to 13.4 % CO
oil with flame retention head
burners
, 11 ppm to
2
values) 8.6 % CO
gas, 9.8 % CO
11 % CO
Set up burners and air or
•
for LP gas, and
2
for #2 fuel oil.
2
water flow volumes to manufacturer’s specifications. You
may need to consult the product service manual or contact
the manufacturer directly for
excess air and CO values.
In the absence of manufactur-
•
er’s excess air specifications,
the following abbreviated
guidelines may be used as
typical values.
Excluding the laws of phys-
•
(4)
ics, every rule has exceptions.
Packaged rooftop gas equipment may have unexpectedly
high O
(14 % O2, or 4 % CO2
2
natural gas) in order to control
side vented vent temperatures
(flame temperature is inversely
proportional to excess air
quantities).
(5)
Variable capacity
burners can fire at over 10 %
O
at low fire. Older 1725 RPM
2
burners without flame retention heads may have a setup
range of 8 % to 10 % CO
Do not try to force a burner
to operate at typical values if
it does not want to willingly
comply.
As in any diagnostic proce-
•
dure, no one thing is used as a
stand-alone proof. We use our
senses, instruments, knowledge and common sense to
gather as much information as
we can in order to support our
2
for natural gas
2
for LP gas
2
for natural gas
2
for LP gas
2
for #2 fuel
2
for natural
2
.
2
theory that a process is, or is
not, functioning as it should.
If a natural gas appliance is
•
installed that requires field
conversion to LP gas, do not
operate the appliance until the
conversion has been completed. If equipment set up for
operation with natural gas is
operated with LP gas, abundant CO will be produced,
sooting is likely, and more
than one kind of headache
will be created. If the LP conversion cannot be done at the
time of installation, disable
the appliance so it cannot be
operated until the conversion
has been completed.
The equipment, components
•
and passageways must be
clean. Components must be in
good mechanical condition,
physically and operationally,
and properly aligned. High CO
in the vent products usually
points to shortcomings of one
or more of those factors.
- Heat exchanger passages
must be clean and clear.
Look for metal flaking, scale
or soot especially at reduced
dimensions around baffles.
- Don’t overlook the importance of injection (Bunsen
type) burner venturi cleanliness. If the interior of the
burner accumulates lint,
develops rust or scale, or
is coated with anything
3 Fluke Corporation Carbon monoxide: A mechanic’s approach
Page 4
that creates friction in the
venturi, then less than
normal primary air will be
entrained and a dirty fire
will develop that secondary
air cannot clean up.
- Cold oil from an outdoor
tank will cause increased oil
flow through an oil nozzle
and resists atomization
and proper fuel/air mixing.
Reduce nozzle size and
increase pump pressure to
clean up the fire.
CO that steadily increases as
•
the burner operates usually
indicates combustion air and/
or venting problems.
The furnace heat
exchanger
The culprit is always a leaking
heat exchanger, and leaking
heat exchangers are deadly. The
firefighters said so. But we know
it ain’t necessarily so. We know
that restricted (and non-leaking)
heat exchanger passages draw
in less secondary air and elevate
CO levels. We know that if the
passageways in a non-leaking
heat exchanger are restricted
enough, then floating flames,
increased CO and roll-out
can occur. But unless a heat
exchanger leak interferes with
combustion or draft, the production of CO is likely to remain
unchanged.
Even a new heat exchanger
can leak. One of the requirements for heat exchanger
manufacture stipulates in ANSI
standard Z21.47 that a new furnace must have a leakage rate
of no more than 2 percent of the
total volume of flue gases. It may
leak when it’s new. Still, heat
exchanger integrity is something
that we try to monitor on an
annual basis. Visual confirmation of a heat exchanger leak is
usually one of our more difficult
tasks. Before breaking out the
visual inspection tools, or tearing down the furnace, try some
old tricks:
Close supply registers to
•
increase static pressure.
If there are no significant
changes before and after the
blower starts, then the heat
exchanger is probably fine.
Check O2 (or CO2) before
•
and after the blower starts.
Increased O
CO
) would indicate air leak-
2
(or decreased
2
age into the heat exchanger.
With oil burners, compare
•
stack O
O
to overfire O2. Higher
2
(lower CO2) in the stack
2
than overfire points toward a
leaking heat exchanger.
With direct-vent products,
•
tee into the more positive
(“+”) pressure sensing tubes
and see if there is a pres-
sure change when the blower
starts. A change in pressure
differential can indicate a heat
exchanger leak.
What not to rely on:
Don’t use a CO reading at a
•
supply register to determine
heat exchanger integrity. That
only proves that the blower is
operating. But two CO readings that show a difference
between the return at the
furnace (provided there are no
return leaks) and the supply
could point to a leaking heat
exchanger.
Don’t rely on a change in CO
•
readings in the vent when the
blower starts up to determine
a leaking heat exchanger. A
leak in the heat exchanger
might not have any effect on
the quality of combustion and
CO production.
Combustion air
Increasing CO and decreasing
O
(or increasing CO2) levels in
2
the vent point to combustion air
problems and possible venting
problems.
We know that adequate
combustion air must always be
available for the operation of
fossil fuel products. Just because
provisions are made that meet
code requirements doesn’t
mean combustion air is indeed
adequate. Even spaces that are
not considered confined (greater
than 50 cubic feet per 1,000
BTU input) still may not have
adequate infiltration for combustion air and venting.
We think nothing of creating access holes in ductwork
for air measurements (dry and
wet bulb temperature, velocity
readings), but before we allow
fossil fuel products to operate,
shouldn’t we make a hole in the
building envelope to verify that
we are not depressurizing the
interior space?
All we need is a 3/16” or
¼” hole to connect to the “+”
side of our micro-manometer
(Fluke 922 Airflow Meter) for
checking pressure differential
between the equipment room
and outdoors. Start all products that vent to the outdoors:
furnaces, boilers, water heaters, exhaust fans, clothes driers.
4 Fluke Corporation Carbon monoxide: A mechanic’s approach
Page 5
Close doors to the equipment
room. Monitor the pressure differential between the equipment
room and outdoors. Then open
the equipment room door (if it is
an interior door) and monitor the
pressure differential again. Any
indoor depressurization means
you need more combustion air.
Venting
Increasing CO and decreasing
O
(or increasing CO2) levels in
2
the vent indicate combustion air
problems and possible venting
problems. Most of the efforts to
ensure adequate combustion air
will apply to venting as well.
But sometimes proper venting
requires more than the provisions that were provided for
combustion air.
Whatever leaves the equipment room through venting to
the outdoors must be replaced
by air from the outdoors. In
the case of Category I vents,
proper draft pressure is very low
(-0.01” to -0.03” wc) and easily
overcome by opposing forces
such as return-duct leaks. Too
much draft is also a problem
because it can reduce product
efficiency, lower vent temperatures and increase building
depressurization.
Vent design and make up air
must be carefully considered in
order to ensure proper venting
under all possible conditions.
Vent gasses that spill from a
draft diverter can ultimately lead
to insufficient combustion air
and high CO production. Combustion products from carelessly
placed sidewall vents can be
drawn into the occupied space
due to building depressurization,
or cause damaging condensation
on structural surfaces or other
equipment and cause operational problems with meters and
regulators located in the vicinity.
Category I venting
Use a draft gauge or your
•
Fluke 922 Airflow Meter to
check draft.
If a cold vent with the appli-
•
ance off can create a -0.03”
wc draft, then the vent is
probably adequate.
Operating draft should be
•
-0.01” wc for both natural
draft and fan assisted products. Oil burners may require
-0.03” wc draft to overcome
heat exchanger internal
pressure drops in order to
maintain -0.01” wc overfire
draft. Operating drafts over
-0.03” wc should be avoided.
A poorly venting chimney will
•
spill flue products from the
draft hood or draft diverter.
CO
is slightly heavier than
2
air and, as concentrations
increase in the still air of the
equipment room, will settle
at floor level where the water
heater burner is located.
The burner is the first to be
starved for combustion air and
produce high levels of CO.
If draft is poor, here are some
•
key reminders:
– Seal any and all return side
duct leaks and filter access
leaks.
– Add a small register in the
supply plenum to overcome equipment room
depressurization.
– Get as much vertical height
off the flue collar as possible
before the first elbow.
– Draft is lost in any vent
section that is less than
vertical. Avoid sloped vent
connectors. Use vertical for
rise, then ¼” per foot rise for
lateral.
– Increase vent connector size
at the flue collar. The code
allows for up to two sizes
of vent connector upsizing
when vent tables allow the
selected size.
– Seal vent leaks such as
leaking cleanouts, poor vent
connector fit in thimble, and
poor thimble fit to clay tile
liner.
– If a masonry chimney has
poor draft, installing a
stainless steel corrugated
reliner will increase friction
and make the draft worse.
Fan-assisted gas equipment
•
is directly coupled to the
vent without any means of
draft control. A dual action
barometric draft control can
be installed and adjusted to
control draft.
Draft hoods, including draft
•
diverters integral to equipment, are vent de-couplers.
This means furnace draft and
chimney draft are separate. A
good vertical vent draft will
vent the equipment room and
usually vent the appliance
also, but an air curtain can be
created at the draft hood and
cause flue gas spillage, even
when the vent draft is good.
When this occurs, replacing
the draft hood with a dualaction barometric draft control
makes the appliance an
integral part of the vent and
allows draft to be controlled—
just like oil burner vents,
except oil vents require only
single-action, not dual-action,
barometric draft controls.
We began thinking about what
characteristics of the system can
affect the quality of combustion.
In closing, maybe we should
think about what characteristics
of the system cannot affect the
quality of combustion? Which
list is longer?
5 Fluke Corporation Carbon monoxide: A mechanic’s approach
Page 6
Combustion air tips
Any return ducts located in
the same space as the fossil
fuel equipment must be sealed
airtight.
Filter access panels must be
sealed airtight.
Returns and filter access
panels located in garages must
be sealed airtight, airtight,
airtight.
When combustion air openings to the outdoors are used,
always use two: one high, one
low. If ducts are used to connect
to the outdoors, use two ducts,
one for each opening. If only one
combustion air supply or duct is
used, it may act like a vent in
the space and try to remove air,
rather than make it available.
Add a small register to the
supply trunk in the equipment
room and make sure there are
no return leaks. This will make
available in the equipment room
the same air that’s available to
the entire structure. Supplying 33 cfm per 100,000 BTU
input allows for 50 % excess air.
When natural draft venting is
used, double this amount. While
you’re at it, add the poor man’s
ventilation system. A duct with
a balancing damper between a
hood installed outdoors and the
return duct will ensure an “on
demand” pressurizing wholehouse ventilation system.
Remember that any outdoor
air access must not use a screen
covering with less than ¼”
mesh. Smaller mesh screens can
become lint traps which block
the supply of outdoor air.
Various after-market combustion air systems are available for
quick and easy installation.
A heat recovery ventilator
could satisfy the requirement
for combustion air, along with
providing healthy indoor air.
See how much money is
available for ensuring adequate
ventilation, venting, and combustion air; then see how much
is left over for the furnace or
boiler purchase. If we didn’t
think our brakes would work,
we wouldn’t even want to start
our car. Don’t even think about
starting a fossil fuel product
until you’re sure that there is
adequate ventilation air.
One would think that direct
vent gas and oil products would
not have combustion air problems. Yet improper design and
installation or poor location can
cause exhaust gas recirculation
into the air intake and create
elevated CO and even sooting.
Chances of lingering direct-vent
exhaust gasses increase with
low velocity discharge, mild
weather, dampness, low barometric pressure conditions, or
any combination thereof. Follow
the manufacturer’s direct-vent
application and installation
requirements very carefully and
don’t use one manufacturer’s
guidelines for another manufacturer’s product.
Footnotes
(1) LEED-EB IEQ
LEED = Leadership in Energy Efficiency and Design
EB = Existing Building
IEQ = Indoor Environmental Quality
(2) Rate of CO2 generation per person = 0.0084 cfm x met rate
Resting met rate = 1.0
Light activity met rate = 1.2
Cooking met rate = 1.8
House cleaning met rate = 2.7
Calisthenics met rate = 3.5
Basketball met rate = 6.3
(3) Calculating CO Air Free
(Oxygen in Air/(Oxygen in Air - Oxygen in Vent Products)) x
CO in Vent Products
(21/(21 - O
(Ultimate CO
Natural Gas: (12/(12 - CO
LPG: (13.7 / (13.7 - CO
#2 Fuel Oil: (15.6/(15.6 – CO
example: (21 % O2/(21 % O2 – 6 % O2 Sample)) x 100 ppm
O
2
CO Sample = 140 ppm CO Air Free
CO
x 100 ppm CO Sample = 124 ppm CO
(4) Values based on National Comfort Institute recommendations
developed from many years of research data
(5) Gas Engineers Handbook
6 Fluke Corporation Carbon monoxide: A mechanic’s approach
Sample)) x CO Sample
2
/(Ultimate CO2 - CO2 Sample)) x CO Sample
2
example: (15.6 % CO2/15.6 % CO2 – 12.5 % CO2 Sample))
2
Sample)) x CO Sample
2
Sample)) x CO Sample
2
Sample)) x CO Sample
2
Fluke. Keeping your world
up and running.
Fluke Corporation
PO Box 9090, Everett, WA 98206 U.S.A.
Fluke Europe B.V.
PO Box 1186, 5602 BD
Eindhoven, The Netherlands
For more information call:
In the U.S.A. (800) 443-5853 or
Fax (425) 446-5116
In Europe/M-East/Africa +31 (0) 40 2675 200 or
Fax +31 (0) 40 2675 222
In Canada (800)-36-FLUKE or
Fax (905) 890-6866
From other countries +1 (425) 446-5500 or
Fax +1 (425) 446-5116
Web access: http://www.fluke.com
©2010 Fluke Corporation.
Specifications subject to change without notice.
Printed in U.S.A. 1/2010 3611250A A-EN-N
Modification of this document is not permitted
without written permission from Fluke Corporation.
®
Loading...