ASHRAE Journal - May 2022 - 48

TECHNICAL FEATURE
concentrations of contaminants than would be calculated
assuming good mixing. In other words, an
individual in the space near the exhaust of these appliances
may be subject to higher concentrations of all the
contaminants of concern. This effect is not, however,
accounted for in this article.
The assumption is that the fuel being burned is
methane. Natural gas is a mixture of fuels heavy in
methane and many of these appliances are designed to
burn propane. In the " Propane " sidebar, these impacts
are discussed in more detail, but the only significant
change is a small increase in carbon dioxide production.
The result of that analysis is that the acceptable
normalized capacity with respect to CO2 is reduced from
0.47 Btu/h·ft3 (4.9 W/m3) for methane to 0.40 Btu/h·ft3
(4.1 W/m3) for propane, which makes carbon dioxide
(from propane) more restrictive than nitrogen dioxide,
but not by much.
The capacity limitations based on carbon dioxide and
nitrogen dioxide are the most restrictive compared
to the other two contaminants. Given the uncertainties
in the calculation, those limitations are practically
the same and to one significant figure can be called
0.4 Btu/h·ft3 (4 W/m3).
As mentioned earlier, improvements in NO2 emission
reductions have happened steadily over time and
no reason exists to assume that could not continue.
Our current results show that limitations based on CO2
and NO2 are not that different, but if NO2 reduction
took place, CO2 would be the limiting contaminant of
concern. Ventilating using CO2 control is a well-known
technology and could be evaluated, which is shown in
the sidebar " Increasing Ventilation. "
Our analysis, as stated earlier, is based on considering
steady state for a well-mixed home. In reality, nothing
is ever perfectly mixed or steady state, and it is worth
noting qualitatively how the situation might change if
one had enough information to do a dynamic calculation
for every home. Most obviously, the concentration
of the contaminants would vary around the steady-state
value we have calculated-with the contaminant concentrations
sometimes being higher and sometimes being
lower. Higher peak values could lead to more conservative
conclusions if instantaneous contaminant levels
were of critical importance, but in general they are not.
There are four issues to consider when looking at
dynamic response. The first, as discussed earlier, is
48
ASHRAE JOURNAL ashrae.o rg M AY 2022
that the appliance must be running long enough to add
a meaningful quantity of contaminants. Our analysis
does not apply to an appliance that is only used for a few
minutes a day, but it is also not an interesting (or typical)
pattern.
The second issue it that the appliance might be cycling,
which would lead to oscillating contaminant values
during the time the appliance was in use. To get a more
accurate estimate of the exposure during that time, it
would be appropriate to use the average emission rates.
Thus, the capacities we refer to in our graphs and conclusions
are best understood as the average capacity
during the time the appliance is cycling.
The third issue is that the air change rate we have used
from 62.2-2019 is not required by that standard to be at
that level for every hour of the year. The standard allows
various approaches that can lead to hours when the air
change may be below the standard level and hours when
it is above it. For most homes just meeting the standard,
the annual average will actually be below the value we
used, but these appliances are more likely to be used
when infiltration is higher. The standard intends the
annual exposure to be equivalent, but there can certainly
be periods when operation of an appliance could
lead to higher or lower levels than we have calculated.
The fourth issue is that our steady-state assumption
applies not only temporally, but spatially. That is, we
have assumed the home is well-mixed. Because the
appliance is a point source, and mixing is never perfect,
the concentrations of contaminants in areas near the
appliance are likely going to be at higher levels than
more remote areas.
Taken together these issues imply that the reader
should consider the specific use case for which they are
using our results. There are times when a more conservative
(or liberal) approach may be justified, but as
a general approach the steady-state assumption is the
place to start.
Conclusions and Recommendations
Unvented combustion produces a variety of potentially
hazardous contaminants, which can have a significant
impact on acceptable indoor air quality. The pollutants
analyzed in detail are carbon dioxide, water vapor, carbon
monoxide and nitrogen dioxide. Two of them, CO2 and
NO2, are likely to be the critical contaminants of concern
when determining acceptability, but moisture impacts
http://ashrae.org
ASHRAE Journal - May 2022

Table of Contents for the Digital Edition of ASHRAE Journal - May 2022
