ASHRAE Journal - November 2019 - 78

COLUMN IEQ APPLICATIONS

from some of the feedback functions mentioned above
is that bypass will always make the efficiency of the filter
smaller, although the absolute impact is bigger for high
efficiency filters because they already have higher efficiencies to start with.
Given the above considerations, we got in the habit
of using the term filtration system efficiency because
filter efficiency is not a just a function of the filter itself.
If we put all these system factors together, Table 1 shows
the new and final effective filter efficiency (mean ± one
standard deviation) of the filter efficiencies for PM2.5 in
all of the homes. Three things to note: the filters do have
higher efficiencies as their MERV goes up, the efficiencies for the electret filters gets lower as they age, and the
standard deviations for all efficiency measurements are
all large relative to the means.
Efficiency is important, but what we really care about
is the overall performance of the filter. One way of measuring performance is to consider the effectiveness of
the filter. Effectiveness considers the loss of particles in
the broader context of particle losses by deposition and
ventilation, and to be effective, a filter has to remove
particles that wouldn't otherwise be removed by ventilation and deposition. We measured the effectiveness
at the same times that we measured efficiency and, as
with filtration system efficiency, we found a wide variety
of effectivenesses that ranged from near zero to almost
80%. Median effectiveness when the system was running
ranged from 15% to 45% for all filters. Effectiveness was
even less dependent on filter MERV than efficiency. This
makes sense because most of our sample were relatively
old and leaky homes and thus filters were adding relatively little to already high loss rates from exfiltration.
The short-term measurement of effectiveness is
an overstatement of actual concentration reductions
because most residential systems do not operate continuously. Instead, they cycle on and off as the thermostat calls for heat or cooling. Typical North American
residential runtimes are a little under 20% when averaged over the whole year, but vary a lot on a day-to-day
basis based on outdoor temperatures. This means that
even if you have a filter that has a high filtration system
efficiency, it doesn't matter most of the time because air
is not going through it. In our sample, runtimes were
even lower with a median runtime over the entire study
of 10% (the study year was a relatively mild year in both
winter and summer). We did have some homes that ran
78

ASHRAE JOURNAL

ashrae.org

N O V E M B E R 2 0 19

TABLE 1 Mean (standard deviation) PM2.5 effective filter efficiencies for new and
used filters.
MERV 8

MERV 8E

MERV 11E

MERV 14E

New

16 (15)

38 (13)

41 (17)

59 (18)

Used

19 (19)

26 (21)

26 (18)

48 (17)

their systems in fan only mode some of the time (mostly
for mixing purposes as older Toronto homes can get
quite thermally stratified), but most only saw any potential benefit of filtration when their thermostat called for
heating or cooling.
Beyond any measurements of efficiency or effectiveness, what we really care about is whether there was
any difference in particle concentrations with different levels of filtration in the homes. To evaluate this,
we relied on low-cost monitors that measured the
particle concentration for >0.5 and >2.5 µm particles
every hour in each of the homes. The short summary
is that there wasn't any statistical difference in particle
concentrations between the times that the four filters
were installed. The longer answer is that there was a lot
going on in these homes besides filtration. They had the
usual sources and had elevations of particles either from
increases in outdoor particles or from indoor sources
ranging from 30 to 60% of the time. And, as mentioned
previously, air wasn't going through the filter most of the
time. This highlights the fact that filters are one part of
the complex residential environment and their performance can't be extrapolated from laboratory tests alone.
The central question is what to do with this information. One approach would be to develop a residential
filter laboratory standard similar to existing filter
standards but geared toward filters used in residential
systems. Although this is appealing in some ways, ultimately, I don't think that such a standard addresses the
fundamental system issues that drive filtration when
installed in actual homes. In-situ performance testing
would be ideal, but there isn't an easy way to standardize such testing in an economical manner. What may be
more useful in the long term is to provide much better
information to homeowners about how to get the most
possible benefit from a filter (e.g., install a gasketed filter slot and run fans more than needed for conditioning,
change the filter more frequently). Ultimately the issue
touches on several areas of expertise within ASHRAE:
building enclosures, air distribution, source management, ventilation, duct design, and fans.

ASHRAE Journal - November 2019

Table of Contents for the Digital Edition of ASHRAE Journal - November 2019