Engineering Report: H242

Engineering Services
FAN 351

Johnson Controls, Inc.

Controls Group

507 E. Michigan Street AIR FILTRATION

P.O. Box 423
Milwaukee, WI 53201

CONTENTS
THEORY OF AIR FILTRATION................................................................................. 2
Electrostatic Attraction .......................................................................................... 2
Straining................................................................................................................ 2
Diffusion................................................................................................................ 3
Impaction .............................................................................................................. 3
Sizes of Airborne Contaminants ........................................................................... 5
FILTER RATING METHODS ..................................................................................... 6
Arrestance and Efficiency Tests ........................................................................... 6
ASHRAE Weight Arrestance Test ................................................................... 6
ASHRAE Atmospheric Dust Spot Efficiency Test............................................ 7
DOP Efficiency Test ........................................................................................ 7
Dust-Holding Capacity Tests ................................................................................ 8
TYPES OF AIR CLEANERS...................................................................................... 9
Fibrous Media Unit Filters..................................................................................... 9
Viscous Impingement Filters ........................................................................... 9
Sinuous Media ......................................................................................... 10
Formed-Screen Media ............................................................................. 10
Random Fiber Media ............................................................................... 10
Dry Type Air Filters........................................................................................ 12
Renewable Media Filters .................................................................................... 15
Moving-Curtain Viscous Impingement Filters ................................................ 15
Moving-Curtain Dry Media Filters.................................................................. 16
Performance of Renewable Media Filters ..................................................... 17
Electronic Air Cleaners ....................................................................................... 18
Ionizing Type Electronic Air Cleaners............................................................ 18
Space Charge ............................................................................................... 19
Charged-Media Electronic Air Cleaners ........................................................ 20
SELECTION AND MAINTENANCE......................................................................... 20
FILTER INSTALLATION ......................................................................................... 23
SAFETY REQUIREMENTS ..................................................................................... 25
GLOSSARY OF TERMS.......................................................................................... 25

3/72 1
 Engineering Report: H242

THEORY OF AIR FILTRATION

Deposition of airborne contaminants on filter media occurs by four methods:

1. Electrostatic Attraction
2. Straining
3. Diffusion
4. Impaction

Electrostatic Attraction

Dust picks up an electrical charge as it passes through the air in the same manner
that a person picks up a static charge by walking on a rug. These particles, in turn,
tend to stick to the filter fibers which have also picked up a charge. Synthetic fibers
are supposed to pick up a charge easily, but in practice, no media seems to work
effectively by electrical attraction without utilizing an induced electrical field to
provide the charge. This effect can be readily demonstrated on a desk top, but
actual filter results do not show a practical effect.

Figure 1 shows the most common form of electronic air cleaner, in which the
airborne particles are charged by means of a power supply in the first section and
collected in the following section. This type of air cleaner can be made highly
effective.

Figure 1. Electrostatic Attraction

Straining

Since all filters remove particles much smaller than the openings in the filter,
removal of particles by straining is not a major factor in most filters. It does,
however, account for filtering of lint and other large particles.

Figure 2. Straining

3/72 2
 Engineering Report: H242

Diffusion

An air suspension of very fine dust particles will behave much like a true gas; that is,
the particles will tend to diffuse from areas of high concentration to areas of low
concentration in the same manner that cigarette smoke will tend to diffuse
throughout a room.

Once a particle has been attracted to a filter fiber, the concentration of dust particles
in the vicinity of the fiber has been reduced. This reduction in concentration causes
the other particles to migrate to the surface of the fiber where the density of airborne
particles is zero.

Diffusion works best with fine filter fibers and at low air velocities. At high air
velocities, the particles are carried beyond the fibers before they have a chance to
diffuse in the same manner that a fan would remove smoke from a cigarette before
it could spread throughout a room.

Figure 3. Diffusion

Impaction

As air flows through a filter, it must bend or change direction many times to flow
around the filter fibers. Because of their inertia, particles resist change in direction
and attempt to continue on in their original paths. For this reason, they collide with
and adhere to the filter fibers regardless of the changing path of the air.

If the air is moving too slowly in relation to the direction changes through the media,
the dust will change direction an be carried through the filter with the air. It is for this
reason that most filters have a minimum as well as a maximum velocity rating.

Figure 4. Impaction

3/72 3
 Engineering Report: H242

Summary

For media filters, efficiency is determined by the relative sizes of the filter fiber, and
its packing density, as well as the adhesive used on the fiber, the size and character
of the dust particles, and the velocity at which the air is moving. Generally, large
fibers are effective on large particles at high velocities and smaller fibers work best
on smaller particles at lower velocities.

If the filter fibers are designed to remove only large particles, they will not be
effective on smaller diameter particles regardless of the depth of the filter. For this
reason, smaller diameter fibers must be used to stop fine dust particles. Also, many
filter fibers will initially stop dust particles, but may release them later unless an
adhesive is employed to permanently trap the particles.

Figure 5 shows the wide range of contaminants found in air and the sizes
associated with the various contaminants. The micron () is the commonly used
measurement unit. One micron is one millionth of a meter or 0.000039-inch. The
human eye can distinguish single particles about 20 microns and larger in diameter.

3/72 4
 Engineering Report: H242

Figure 5. Sizes of Airborne Contaminants

3/72 5
 Engineering Report: H242

FILTERING RATING METHODS

Ideally, a filter test method should simulate the performance of the filter when in
actual operation, as well as provide a means of comparing the performance of
different types of filters. Because of the variation in concentration and composition
of airborne contaminants from area to area, and from time to time in a single area,
no accelerated testing procedure can perfectly duplicate the conditions under which
the filter will actually operate. However, the appropriate testing procedure will
provide a good indication of the comparative performance of various filters.

For general ventilation air use, ASHRAE Standard No. 52-68, Methods of Testing
Air Cleaning Devices, represents the most recent and most meaningful methods.

Filter rating methods are of two kinds:

1. Arrestance and Efficiency Tests.

2. Dust Holding Capacity Tests.

Arrestance and Efficiency Tests

1. ASHRAE Weight Arrestance Test.

Procedure

An artificial contaminant consisting of 72 percent standardized air cleaner dust,

fine; 23 percent Molocco Black; and five percent No. 7 cotton linters is fed into
the test duct at a controlled rate. As each increment of dust is fed, the filter is
weighed to determine the fraction of dust stopped, and an absolute filter
downstream is weighed to determine the dust passing the filter. The percentage
of dust stopped is the arrestance, as measured by:

Weight of Dust Passing Filter

100 1 -
Weight of Dust Fed to Filter

The procedure is repeated until the manufacturers limits are reached. The
incremental arrestance values are weight-averaged to arrive at an arrestance
rating.

Conclusions

This method is useful for comparing the less efficient impingement type filters.
Since the larger dust particles account for most of the weight of the test dust, this
test will not indicate how effective a filter is in removing the smaller stain-
producing particles.

3/72 6
 Engineering Report: H242

Actually, about 70 percent of the soiling capacity of normal atmospheric air is

represented by only about 7 percent (by weight) of the particles. Moreover, this 7
percent consists of very small particles in the 0.1 to 3.0 micron range.

2. ASHRAE Atmospheric Dust Spot Efficiency

Procedure

To accelerate this test, the filter is loaded with ASHRAE synthetic dust as in Step
1 preceding, but no arrestance measurements need be taken. A white paper
target, mounted in an air sampler, is located in the duct before and after the filter
to be evaluated. After each increment of loading, the synthetic test dust is
stopped and air containing normal atmospheric contaminants is run through the
filter. The percentage of change in the staining (measured photometrically) of the
two paper discs is used to calculate the efficiency. As in the arrestance
determination, a weighted average is used as the rating value after completion of
the required loading.

Conclusions

This test is more indicative of the ability of a filter to remove the small stain-
producing particles in the air, but it cannot be used as an absolute test because
of the variations in atmospheric contaminants from area to area and time to time
in a single location.

3. DOP Efficiency Test

Procedure

The test aerosol is a dioctylphthalate (DOP) mist consisting of particles

approximately 0.3 micron in diameter. The concentration of particles (by count)
measured upstream and downstream of the filter determines the efficiency (light
scattering methods are used to count the particles).

Conclusions

This method is generally used to rate the very high efficiency or absolute filters
because of the small particle size of the contaminant and the very high
efficiencies usually demanded (99+ percent). The DOP test is usually applied
only to filters for clean rooms or other special applications requiring very clean
or sterile air.

3/72 7
 Engineering Report: H242

Dust Holding Capacity Tests

The capacity of the filter to hold dust is an important characteristic which determines
the effective field life of the filter. To be meaningful for comparing filters, the air
velocity, dust held, the arrestance values and the pressure drop of the filters must
all be related. The ASHRAE Standard 52-68 outlines the detailed procedure. The
dust holding capacity of a filter is the dust held (in grams) after the manufacturers
stated pressure drop is reached.

Table 1
Filter Testing Methods
Method of Method of Method of Class of filters
evaluating evaluating evaluating for which test
Test Test inlet dust outlet dust filter method is most
Method Dust concentration concentration efficiency applicable

ASHRAE Synthetic Dust Pre-weighed All cleaned air Percent retention Viscous-coated
Weight composed of 72% charge of dust fed passed through a by weight impingement-type
Arrestance Standardized Air to test filter pre-weighed filter ventilation air
Cleaner dust-fine, capable of filters and other
23% Molocco retaining dust low and medium
black, 5% No. 7 penetrating filter effectiveness
cotton liners undergoing test filters

ASHRAE Natural Staining or Same as inlet Percent increase Electrostatic

Atmospheric Dust Atmospheric air discoloration of a concentration in discoloration precipitators and
Spot white filter paper properties with fibrous air filters
increasing degree capable of
of dust loading effecting
substantial
reduction in
staining properties
of ventilation air

DOP Droplets of Optical Same as inlet Percent by HEPA-type filters

dioctylphthalate measurement of concentration number of droplets and those
0.3 micron in size amount of light penetrating filter possessing similar
scattered by DOP filtration
characteristics for
special
applications

3/72 8
 Engineering Report: H242

TYPES OF AIR CLEANERS

All the common types of air cleaners fall into three broad categories:

1. Fibrous media unit filters in which accumulating dust load causes pressure drop
to increase continuously to the end of an operating period determined by
maximum permissible pressure drop. At the end of this period, filters are either
replaced or washed for re-use. This category includes both viscous impingement
and dry type air filters.

2. Renewable media filters in which fresh media is introduced into the air stream as
needed to maintain essential constant resistance.

3. Electronic air cleaners where the pressure drop is essentially constant.

Various combinations of the above types are used. For example, an electronic air
cleaner may be used as an agglomerator with a fibrous media downstream to catch
the agglomerated particles blown off the plates. Also, a renewable media filter may
be used upstream of a high efficiency unit filter to extend its life.

There is also a charged media filter in which particle deposition on media fibers is
increased by the use of an electrostatic field. In this case, the pressure loss
increases like a fibrous media filter. Humidities of 60 percent or greater are likely to
reduce efficiencies. This type of filter is not in the same class as the one shown in
Figure 1.

FIBROUS MEDIA UNIT FILTERS

Viscous Impingement Filters

These are flat panel filters made up of coarse fibers and have a high porosity. The
filter media is coated with a viscous substance, such as oil, which acts as an
adhesive on particles that impinge on the fibers; hence, the name viscous
impingement. Design air velocity through the media is usually in the range of 250 to
700 fpm. These filters are characterized by low pressure drop, low cost, good
efficiency on lint but low efficiency on normal atmospheric dust. They are commonly
made 1/2 to 4-inches thick with one and two-inches nominal thickness being most
popular. In the thicker configurations, the dust-holding capacity is substantial. Unit
panels are available in standard and special sizes up to about 24 x 24 inches. This
type of filter is often used as a pre-filter to higher efficiency filters.

3/72 9
 Engineering Report: H242

A number of different materials have been used as the filtering medium including
coarse (15 to 60 micron diameter) glass fibers, animal hair, vegetable fibers,
synthetic fibers, metallic wools, expanded metals and foils, crimped screens,
random-matted wire, and synthetic open-cell foams. The arrangement of the
medium in this type of filter involves three basic configurations:

1. SINUOUS MEDIA. The filtering medium (consisting of corrugated metal or

screen strips) is held more or less parallel to the air flow. The direction of air flow
is forced to change rapidly in passing through the filter, thus giving inertial
impingement of dust on the metal elements. This type of filter design generally
does not have any recommended direction of air flow.

2. FORMED-SCREEN MEDIA. Here, the filter media (screens or expanded metal)

is crimped to produce a high-porosity media that avoids collapsing. Air flows
through the media elements, and dust impinges on the wires. The relatively open
structure allows the filter to store substantial quantities of dust and lint without
plugging.

3. RANDOM FIBER MEDIA. Fibers, either with or without bonding material, are
formed into mats of high porosity. Media of this type are often designed with
fibers packed more densely on the leaving air side than on the entering air side.
This arrangement permits the accumulation of larger particles and lint near the
air-entering face of the filter, and the filtration of finer particles on the more
closely packed air-leaving face. Fiber diameters may also be graded from coarse
to fine from air-entering to air-leaving face. The fibers used and the methods of
formation often are cheap enough to allow such filters to be thrown away when
clogged with dirt, rather than being cleaned and reused, as is common practice
with the other two types.

Although viscous impingement filters usually operate in the range of 300 to 600 fpm,
it is possible to operate at higher velocities. The limiting factor other than increased
flow resistance is the danger of blowing off agglomerates of collected dust and the
viscous coating on the filter as well.

In general, unit filters are serviced when their operating resistance reaches 0.5-inch
of water. The decline in filter efficiency which occurs when all the viscous coating
has been absorbed by the collected dust, rather than the increased resistance due
to dust load, may be the limiting factor in operating life.

3/72 10
 Engineering Report: H242

The adhesive or oil used on a viscous-impingement type filter requires careful

engineering. Filter efficiency and dust-holding capacity are both dependent on the
specific type and quantity of adhesive used; this information is an essential part of
test data and filter specifications. Desirable characteristics of an adhesive, in
addition to its ability to deliver the desired efficiency and dust-holding capacity are:

A. A low percentage of volatiles to prevent excessive evaporation.

B. A viscosity that varies only slightly within service temperature range.

C. The ability to inhibit growth of bacteria and mold spores. (An adhesive material
must not be a nutrient for such particles, of course. On the other hand,
bactericidal agents added to adhesives do not show practical value in rendering
dirty filters sterile.)

D. A high capillarity, or the ability to wet and retain the dust particles.

E. A high flash point and fire point.

F. Freedom from odor.

Table 2
Performance of
Unit Viscous-Impingement Filters

ASHRAE ASHRAE
ASHRAE Atmospheric Dust-Holding
Filter Weight Dust Spot Capacity,
Thickness, Arrestance, Efficiency, Grams per
Inches Percent Percent 1000 cfm Cell

>1 20 - 50 15-10 170 - 140

1 - 1-3/4 50 - 75 15-15 120 - 360
1-3/4 - 2-1/2 60 - 80 15-20 180 - 540
2-1/2 - 4 70 - 85 10-25 240 - 760

Typical performance of viscous-impingement unit filters operating within typical

resistance limits is shown in Table 2.

3/72 11
 Engineering Report: H242

Dry Type Air Filters

The media used in dry type air filters are random fiber mats or blankets of varying
thicknesses, fiber sizes, and densities. Media of bonded glass fiber, cellulose fibers,
wool felt, asbestos, synthetics, and other materials have been used commercially.
The medium in filters of this class is frequently supported by a wire frame in the
form of pockets or V-shaped pleats. In other designs, the media may be
selfsupporting because of inherent rigidity or because air flow inflates it into
extended form. Pleating of the media provides a high ratio of media area to face
area, thus allowing reasonable pressure drop in spite of the density and fineness of
the media.

In some designs of dry type air filters, the filter media is replaceable and is held in
position in permanent metal cell sides. In other designs the entire cell is disposed of
after it has accumulated its dirt load.

The absence of an adhesive on the filter medium does not necessarily qualify it as a
dry type air filter. The distinguishing characteristics are the generally higher
efficiency produced by the smaller size fibers and the fact that the fibers are spaced
comparatively close together to obtain a dense mat or pad. A low resistance pad
consisting of large fibers having no adhesive coating would be a dry filter but would
be less effective than a typical viscous-impingement filter.

3/72 12
 Engineering Report: H242

TABLE 3
PERFORMANCE OF DRY MEDIA FILTERS

ASHRAE ASHRAE
ASHRAE Atmospheric Dust-Holding
Filter Media Type Weight Dust Spot DOP Capacity
Arrestance, Efficiency, Efficiency, Grams per
Percent Percent Percent 1000 cfm Cell

Finer open cell foams and textile

denier non-wovens 70 - 80 15 - 30 0 180 - 425

Thin, paper-like mats of glass

fibers, cellulose 80 - 90 20 - 35 0 190 - 180

Mats of glass fiber multi-ply

cellulose, wool felt 85 - 90 25 - 40 5 - 10 190 - 180

Mats of 5 to 10 micron fibers,

1/4 to 1/2 in. thickness 90 - 95 40 - 60 15 - 25 270 - 540

Mats of 3 to 5 micron fibers,

1/4 to 3/4 in. thickness > 95 60 - 80 35 - 40 180 - 450

Mats of 1 to 4 micron fibers,

mixture of various fibers and > 95 80 - 90 50 - 55 180 - 360
asbestos

Mats of 1/2 to 2 micron fibers

(usually glass fibers) NA 90 - 98 75 - 90 190 - 270

Wet laid papers of mostly

submicron glass and asbestos NA NA 95 - 99 .999 500 - 1000
fibers (HEPA filters)

Membrane Filters (membranes of

cellulose acetate, nylon, etc. NA NA 100 NA
having micron size holes)

NA: Indicates that test method cannot be applied to this level of filter.

3/72 13
 Engineering Report: H242

The efficiency of dry type air filters is usually higher than that of viscous-
impingement type filters, and the variety of media available makes it possible to
furnish almost any degree of cleaning efficiency desired. Modern dry type filter
media and filter configurations also give dust-holding capacities generally higher
than viscous-impingement filters. Coarse pre-filters placed ahead of high efficiency
dry type filters may be economically justified by the even longer life they give the
main filters.

Typical performance of some types of filters in this group, when operated within
typical rated resistance limits and over the life of the filters, is shown in Table 3.

The initial resistance of a dry type filter will vary with the choice of media and the
filter geometry. Commercial designs typically have an initial resistance from 0.05 in.
w.g. to 1.0 in. w.g. It is customary to replace the media when the final resistance of
0.5 in. w.g. is reached for the low resistance units and 2.0 in. w.g. for the highest
resistance units. Dry media providing higher orders of cleaning efficiency will offer a
higher resistance to air flow. The operating resistance of the fully dust-loaded filter
must be considered in the system design since that is the maximum resistance
against which the fan will be called upon to operate. If the variation between initial
and final resistance is sufficient to cause a problem in system design, the filters may
be replaced on the rotation system discussed under viscous-impingement unit
filters.

Flat panel filters with media velocity equal to duct velocity are possible only in the
lowest efficiency units of the dry type (open cell foams and textile denier non-woven
media, mats or batting). Initial resistance of this group, at rated air flow, is mainly
between 0.05 and 0.25-inch of water. They are usually operated to a final resistance
of 0.5 to 0.7-inch of water.

On dry filters of the intermediate efficiency ranges, the filter media area is much
greater than the face area of the filter; hence, velocity through the filter media is
substantially lower than the velocity approaching the filter face. Media velocities
range from 5 to 90 fpm although the approach velocities run to 750 fpm. Depth in
direction of air flow varies from 2 to 36-inches.

3/72 14
 Engineering Report: H242

Filter media used in the intermediate efficiency range include those of:

1. Fine glass fibers, 0.7 to 10 microns in diameter, in mat from up to 1/2-inch thick.

2. Thin nonwoven mats of fine glass fibers.

3. Nonwoven mats of comparatively large diameter fibers (approximately 20

microns) in greater thicknessews (up to 2 inches).

Dry filters of very high efficiency, HEPA (High Efficiency Particulate Air) filters, are
made in an extended surface configuration of deep space folds of submicron glass
or glass-asbestos fiber paper. Such filters operate at duct velocities near 250 fpm,
with resistance rising from .10 to 2.0 in. w.g. or more over their service life. These
filters are the standard for clean rooms and nuclear and toxic-particulate
applications.

Membrane filters are used chiefly for air sampling and for specialized small scale
applications where their extreme efficiency compensates for their fragility, high
resistance, and high cost. They are available in many pore diameters and
resistances, and in flat sheet and pleated forms.

RENEWABLE MEDIA FILTERS

Moving-Curtain Viscous Impingement Filters

Automatic moving-curtain viscous filters are available in two main types. In one type,
random-fiber media is furnished in roll form. Fresh media is fed automatically across
the face of the filter, while the dirty media is rewound onto a roll at the bottom. When
the media roll is exhausted, the tail of the media is wound onto the take-up roll, and
the entire roll thrown away. A new roll is then installed, and the cycle is repeated.
Moving-curtain filters have the media automatically advanced by motor drives on
command from a timer or pressure switch control. Pressure switch controls measure
the pressure drop across the media, and switch at chosen upper and lower set
points. This control conserves the media, provided that great care is taken in
locating static pressure tubes so that the true pressure drop is being detected by the
controls. In addition, modulating fresh air and return air dampers can affect air flow
patterns and total flow to such an extent that pressure drop control systems do not
function properly. Timers avoid these problems; their duty cycles can be adjusted to
provide satisfactory operation with acceptable media consumption.

3/72 15
 Engineering Report: H242

In the other type of automatic viscous-impingement filter, chainlinked screening, or

linked perforated plates are installed on a traveling curtain which intermittently
passes through an adhesive reservoir where the media gives up dust load, and at
the same time, takes on a new coating of adhesive. The media thus forms a
continuous curtain which moves up one face and down the other face. The curtain,
being formed of metal and continually cleaned and renewed with fresh adhesive,
lasts the life of the filter mechanism.

Periodically, the precipitated dirt must be removed from the adhesive reservoir. This
is generally done by scraping the dirt into a tray which can be conveniently
suspended from the reservoir lip.

Moving-Curtain Dry-Media Filters

Random-fiber (nonwoven) dry media of relatively high porosity are also used in
moving-curtain (roll) filters, for general ventilation service. Operating duct velocities
are generally lower than for viscous-impingement filters, being in the area of 200
ft./min. Special automatic dry filters are also available which are designed for the
removal of lint in textile mills and dry cleaning establishments and for collection of
lint and ink mist in press rooms. The media used is extremely thin, and serves only
as a base for the buildup of lint, which then acts as a filter medium.

In the previous forms of renewable dry media filters, the dirt-laden media is
discarded when the supply roll is used up.

Another form of filter designed specifically for dry lint removal consists of a moving
curtain of wire screen which is automatically vacuum cleaned at a position out of the
air stream. Recovery of the collected lint is sometimes possible with such a device.

3/72 16
 Engineering Report: H242

Table 4
Performance of Renewable Media Filters (Steady-State Values)

ASHRAE
Type ASHRAE Atmospheric ASHRAE
Description of Weight Dust Spot Dust-Holding Velocity
Media Arrestance, Efficiency, Capacity F PM
Percent Percent g/Sq. Ft.

20 - 40 micron glass and synthetic Viscous

fibers, 2 - 2-1/2 in. thick Imp. 70 - 82 10 - 25 60 - 175 500

Permanent metal media cells or Viscous NA

overlapping elements Imp. 70 - 80 15 - 20 (permanent 500
media)

Coarse textile denier Adhesive

nonwoven mat, 1/2 - 1 in. thick Coated 70 - 80 10 - 20 20 - 50 500

Coarse textile denier

nonwoven mat, 1/2 - 1 in. thick Dry 60 - 80 10 - 20 15 - 70 500

Fine textile denier

nonwoven mat. 1/2 - 1 in. thick Dry 80 - 90 20 - 30 10 - 50 200

NA: Indicates that test method cannot be applied to this level of filter.

Performance of Renewable Media Filters

Dust-holding capacity and efficiency (or arrestance) are equally as important for
renewable media filters as for unit filters. The modified test procedures required for
this type of filter are specified in the ASHRAE Air Filter Test Standard 52 - 68. Test
dust feeding methods and determination of arrestance and efficiency are the same
as for unit filters. However, the test procedure calls for fresh curtain or media to be
advanced into the air stream when resistance reaches the manufacturers design
level, usually 0.4 to 0.5 in. w.g. Cycles of dust feeding and curtain advance are
repeated until approximately two complete duct heights of curtain have been
advanced. Arrestance and efficiency values are obtained at intervals during this
process to establish steady state values. Dust-Holding Capacity in this case is
defined as (steady state arrestance) x (dust loading per unit area of media).
ASHRAE arrestance, efficiency, and dust-holding capacities for typical viscous-
impingement and dry renewable media filters are listed in Table 4.

3/72 17
 Engineering Report: H242

ELECTRONIC AIR CLEANERS

Electronic air cleaners use electrostatic precipitation principles to collect particulate

matter but operate on lower voltages than the type commonly used on industrial
stacks. The designation Electronic Air Cleaner has been standardized to distinguish
the class of electrostatic precipitator which alone is suitable for cleaning ventilating
air. Two types are produced for commercial service, the ionizing type and the
charged-media, non-ionizing type.

Ionizing Type Electronic Air Cleaners

Figure 1 illustrated the operation of this type of filter schematically. Positive ions
generated at the high potential ionizer wire flow across the air stream, striking and
adhering to any dust particles carried by the air stream. These particles then pass
into the system of charged and grounded plates. They are driven to the plates by
the force exerted by the electric field on the charges they carry. The dust particles
which reach the plates are thus removed from the air stream.

In a typical electronic air cleaner design, a D.C. potential of 12,000 volts is used to
create the ionizing field, and 6000 volts is maintained between the plates upon
which the precipitation of dust occurs. These voltages necessitate some safety
measures. A typical arrangement provides means for making the equipment
inoperative when any door affording access to high-voltage parts is opened.
Operation can be resumed only after all doors are closed. The voltages necessary
for operation of the equipment are usually obtained by means of high-voltage direct
current power packs operating from a 110-120 volt, 60-cycle, single-phase building
service. These may have the tube type rectifiers or solid state types employing
silicon or selenium. Electric power consumption is about 12 to 15 watts per 1000
cfm. Air cleaners of this type, unless incorporating pre-filters or after-filters, offer
negligible resistance to air flow and, therefore, care must be exercised in arranging
the duct approaches on the entering and leaving sides of the cleaners in order that
the air flow may be distributed uniformly over the cross-sectional area. The
efficiency of the electronic air cleaner is sensitive to air velocity, and the device itself
has much less tendency to rectify air stream velocity differences than filters which
have higher resistances. In most systems, resistance is deliberately added in the
form of a perfor-ated plate, pre-filters or after-filters, for the purpose of obtaining
uniform distribution of air. The resistance, including the baffles or after-filters, is
relatively constant and generally ranges from 0.15 to 0.25 in. of water at the usual
velocities of 300 to 500 fpm. Such plates or filters, however, cannot compensate for
defects in the duct layout ahead of the cleaner. Suitable screens of no coarser than
16 mesh should be installed across all outdoor air entrances to prevent insects,
leaves, bits of paper, and similar material from entering the cleaner. Where lint is

3/72 18
 Engineering Report: H242

present in appreciable quantities, devices for lint removal should be installed ahead
of the cleaner with provision for cleaning the lint collector periodically.

Electronic air cleaners of the ionizing type are efficient, low-pressure drop devices
for removing fine dust and smoke particles. The ASHRAE atmospheric dust spot
efficiencies range from 70 to above 95 percent. Collector plates are often coated
with a special oil as an adhesive. Cleaning is generally accomplished by washing
the cells in place with hot water from a water hose or by means of fixed or moving
nozzle systems. The bottom of the equipment is made water-tight and provided with
a drain.

Electrical forces drive most particles to the collecting surface but cannot hold them
there. In fact, after a particle touches the collecting surface, the electrical force
reverses and tends to pull it off, and the dust is held only by intermolecular adhesion
forces. It is, therefore, very important with the washed type of electronic air cleaner
to be sure that either the dust is naturally adherent or that the plates are always
covered with adhesive.

Electronic air cleaners, however, are sometimes used without any adhesive
treatment on the plates. Under such conditions, the precipitator forms agglomerates
which eventually blow off the plates. It must therefore be followed downstream by a
secondary filter or storage section. The dry agglomerates produced in the
precipitator are allowed to blow off and be caught by the downstream filter. The use
of an automatic replaceable-media filter for catching these agglomerates results in
an overall combination which provides a high degree of cleaning efficiency, and also
the convenient maintenance associated with an automatic filter of this design.

Another type of electronic cleaner uses a high efficiency dry filter as the collector. In
this version, the ability of electronic air cleaner portion to agglomerate very small
particles is utilized to increase the life of the high efficiency filter, since the weight of
dust held by dry filters increases with increasing particle size of the dust collected.
When handling certain types of atmospheric dust, the ionizing wires and struts may
require periodic cleaning.

Space Charge

Dust which passes through an ionizer and is charged but not removed carries the
electrical charge into the space. If continued on a large scale, this will build up a
space charge in the space which tends to drive this charged dust to the walls. Thus,
and electronic air cleaner which, for any reason, charges but does not remove the
dust can blacken walls faster than if no cleaning devices were used.

3/72 19
 Engineering Report: H242

Charged-Media Electronic Air Cleaners

The charged-media type air cleaner combines certain characteristics of both dry
filters and electronic air cleaners. It consists of dielectric filtering medium, usually
arranged in pleats as in typical dry filters. No ionization means are employed. The
dielectric filtering medium may consist of glass fiber mat, cellulose mat, or other
similar material and is supported on, or in contact with, a gridwork consisting of
alternately grounded and charged members, the latter usually being held at a
potential of several thousand volts D.C. An intense and non-uniform electrostatic
field is thus created through the dielectric medium. Airborne particles approaching
the field are polarized and drawn to filiments or fibers of the media.

A cleaner of this type offers resistance to air flow, when clean, on the order of 0.10
in. of water at 250 fpm velocity. The resistance of the charged-media type electronic
cleaner rises as dust is accumulated on the media. Because of these
characteristics, the filter tends to equalize the air distribution over the face of the
filter. Like typical replaceable-media mechanical filters, the charged-media
precipitator is serviced by replacing the filtering medium. The dielectric properties of
the media are impaired when the relative humidity exceeds 60 to 70 percent. The
efficiency of these filters is usually in the 50% ASHRAE atmospheric dust spot
range.

SELECTION AND MAINTENANCE

To evaluate filters and air cleaners properly for a particular application, three factors
should be carefully weighed:

1. Degree of air cleanliness required.

2. Disposal of the dirt after it is removed from the air.

3. Amount and type of dust in air to be filtered.

These factors affect initial costs, operating costs, and the extent of maintenance that
will be required. Savings, from reduction in housekeeping expenses, protection of
valuable property and equipment, ability to carry on dust-free manufacturing
processes, improved working conditions and even health benefits, should be
credited against the cost of installing and operating an adequate system. The
capacity and physical size of the unit required may emphasize the need for low
maintenance cost. Operating costs, predicted life, and efficiency are as important as
initial cost, because air cleaning is a continuing process.

3/72 20
 Engineering Report: H242

Viscous-impingement unit filters do not have efficiencies as high as can be expected

from dry types of unit filters, but their initial cost and upkeep are generally low. They
require more frequent attention than the moving-curtain type if the resistance is to
be maintained within reasonable limits. If higher efficiencies are required, the dry
type filters or electronic air cleaners should be given consideration.

An advantage of the automatic filters of both the moving-curtain and replenishable-

media type is the lesser amount of attention which they require. Such devices are
therefore recommended for consideration where large volumes of air are being
handled, where the installation is remote, or where reliable and frequent attention to
filters cannot be assumed. The relatively constant pressure drop of this type of filter
is also an advantage. The initial cost is substantially higher than that of unit filters,
but the saving in maintenance cost tends to offset this disadvantage, although, with
age, maintenance may increase substantially due to the added mechanical
complexity.

Developments in extended surface dry type filters, using very fine glass fiber wool
webs or other materials, have brought the efficiency of these filters well into the
range of electronic types. Initial costs of these filters are lower than for electronic
types, but higher than for panel types. Operating and maintenance costs are usually
higher than for panel types and may be higher than for electronic air cleaners.
Pressure drop is greater and slowly increases during its useful life. The advantages
include the fact that no moving parts of mechanical or electrical services are
required. Choice should be based on consideration of both initial costs and
operating costs as well as the degree of cleaning efficiency.

While electronic air cleaners have a higher initial cost, they exhibit high efficiencies
in cleaning atmospheric air, due largely to their ability to remove fine dust which
shows no gravitational effect. System resistance remains unchanged as dirt is being
collected, and the resulting residue is disposed of directly to prepare the equipment
for further duty. The charged-media type electronic air cleaner is less efficient than
the ionizing type when compared on an individual test basis. However, they
incorporate mechanical filtration with electrostatic precipitation, and thus provide the
effectiveness of an intermediate efficiency dry type air filter, should an electrical
outage occur in the filter but not the fan.

Figure 6 shows some applications of various types of filters classified according to

their efficiencies.

3/72 21
 Engineering Report: H242

Figure 6. Typical Applications of Filters Classified by Efficiency

3/72 22
 Engineering Report: H242

FILTER INSTALLATION

Many air cleaners are available in units of convenient size for manual installing,
cleaning, or replacing. A typical unit filter may be 20 or 24 inches square, from one
to several inches thick, and may be either of the dry or viscous-impingement type. In
large systems, the frames in which these units are installed are bolted or riveted
together to form a filter bank. Automatic filters are constructed in sections offering
several choices of width up to six feet, and generally range in height from 3 to 15
feet in 4 to 6 inch increments. Several sections may be bolted together to form a
filter bank.

Several manufacturers provide side-loading filter sections for various types of filters.
Filters are changed from outside the duct, making service areas in the duct
unnecessary and thus saving the cost and space required for these service areas.

The in-service efficiency of an air filter is, of course, sharply reduced if air leaks
around it whether through leaky bypass dampers or poorly designed frames. The
higher the efficiency of a filter, the more attention must be paid to the rigidity and
sealing effectiveness of the frame. In addition, high efficiency filters must be handled
and installed with extreme care.

Air cleaners are commonly installed in the outdoor air intake ducts of buildings, and
generally in the recirculating and bypass air ducts as well. Cleaners are logically
placed ahead of heating or cooling coils and other air conditioning equipment in the
system to protect them from dust. The dust captured by the filters in an air intake
duct is likely to be mostly particulate matter of a greasy nature, while lint may
predominate in dust from within the building.

Where high efficiency filters are employed to protect critical areas such as clean
rooms, it is important that the filters be installed as close to the room as possible to
prevent the pickup of particles between the filters and the outlet. The ultimate in this
trend is the so-called laminar flow room, in which the entire ceiling or one wall
becomes the final filter bank.

The published performance data for all air filters are based on straight-through, un-
restricted air flow. Filters should be installed so that the face area is at right angles
to the air flow whenever possible. Eddy currents and dead air spaces should be
avoided, and air should be distributed uniformly over the entire filter surface, using
baffles or diffusers if necessary. Filters are sometimes damaged if higher than
normal air velocities impinge directory on the face of the filter.

Failure of air filter installations to give satisfactory results can, in most cases, be
traced to faulty installation or improper maintenance or both.

3/72 23
 Engineering Report: H242

The most important requirements of a satisfactory and efficiently operating air filter
installation are:

1. The filter must be of ample size for the amount of air it is expected to handle. An
overload of 10 to 15 percent is regarded as the maximum allowable. When air
volume is subject to future increase, a larger filter should be installed.

2. The filter must be suited to the operating conditions, such as degree of air
cleanliness required, amount of dust in the entering air, type of duty, allowable
pressure drop, operating temperatures, and maintenance facilities.

3. The filter type should be the most economical for the specific application. The
initial cost of the installation should be balanced against efficiency and
depreciation as well as expense and convenience of maintenance.

The following recommendations apply to filters installed with central fan systems:

1. Duct connections to and from the filter should change size or shape gradually to
assure even air distribution over the entire filter area.

2. Sufficient space should be provided in front or behind the filter, or both,

depending on its type, to make it accessible for inspection and service. A
distance of 2 to 3 feet will be required depending on the filter chosen.

3. Access doors of convenient size should be provided to the filter service areas.

4. All doors on the clean-air side should be gasketed to prevent infiltration of

unclean air. All connections and seams of the sheet-metal ducts on the clean-air
side should be as airtight as possible. Any filter bank must be caulked to prevent
bypass of unfiltered air. This is most important when high efficiency filters are
used.

5. Electric lights should be installed in the chamber in front of and behind the air
filter.

6. Filters installed close to an air inlet should be protected from weather by suitable
louvers, in front of which a large mesh wire screen should be provided.

7. Filters, other than electronic air cleaners, should have permanent indicators to
give a warning when the filter resistance reaches too high a value or is
exhausted, as with automatic roll media filters.

8. Electronic air cleaners should have an indicator or alarm system to indicate

when high voltage is off or shorted out.

3/72 24
 Engineering Report: H242

SAFETY REQUIREMENTS

Safety ordinances should be investigated when the installation of an air cleaner of

any considerable size is contemplated. Combustible filtering media may not be
permitted in accordance with most existing local regulations. Combustion of dust
and lint on a filtering medium is possible, though the medium itself may not burn.
This may cause a substantial increase in filter combustibility.

GLOSSARY OF FILTER TERMS

There are certain words and terms which are peculiar to the filter industry. The
following glossary is presented as an aid for training personnel who are new to this
field and as a reference for those who are more experienced. The terms are in
alphabetical order.

Adhesive Substance place in the fibers of the media to aid in trapping

and holding dust particles. Primarily effective on large
particles.

Aerosols Any solid or liquid substance dispersed in air; (see

Contaminants).

Agglomeration Action or process of collecting in a mass; a heap or cluster of

disparate elements.

Air, Outdoor Air taken from outside the air conditioned space or system.

Air, Return Air which has been returned to the plenum from the building
for recirculation or exhaust. A return air duct will generally be
found before the filter media where the return air is mixed with
the incoming outdoor air.

Anemometer Instrument used to measure the velocity of the air.

Apparatus House Boxlike structure which houses the fan, coils, filter bank,
outdoor air dampers and return air dampers. The chambers at
each end are referred to as the plenum chambers.

Approach Velocity see Velocity, Approach.

3/72 25
 Engineering Report: H242

Arrestance and Depending on the type of test used, arrestance is the

Efficiency percentage, by weight, of the contaminants removed from the
air by the filter and efficiency is the percentage of change in
the straining properties of the contaminated air after it has
passed through the filter.

Blinding see Loading, Face.

Blow-Off see Unloading.

Bypass This refers to unfiltered air going around the filter because it
has not been properly sealed in place.

Capacity Volume of air (cfm) which can be delivered through a filter

unit.

Capacity, The gram weight of dust, or contaminants, that is held by the

Dust-Holding filter before the resistance rises to a specified level at a
specified velocity.

Capacity, Rated The cfm rating which is specified by the manufacturer (along
with design velocity).

Casing see Safing.

Centrifugal see Loading, Centrifugal.

Loading

cfm (Cubic Feet Measure of the volume of air being used in a system. An air
Per Minute) handling system rated at 20,000 cfm would have a volume of
air equal to 20,000 cubic feet entering the plenum every
minute.

Coils Unit designed to heat or cool the air after it enters the
building. Normally the coils are located after the filter media
and before the ductwork.

Contaminant Airborne dirt, dust, spores, viruses, bacteria and allergens

which are sometimes referred to as aerosols.

Depth Loading see Loading, Depth.

3/72 26
 Engineering Report: H242

Diffuser System of baffles used to give an even distribution of air flow

through the filter media.

Dust-Holding see Capacity, Dust-Holding.

Capacity

Efficiency see Arrestance and Efficiency.

Face Loading see Loading, Face.

Fiber Breakoff Particles of the media fiber breaking off and entering the air
stream, thereby becoming contaminants.

fpm (Feet Per The speed (velocity) of the air at a given point in the air
Minute) handling system.

Loading, Process by which contaminants tend to collect at the back of

Centrifugal a cube, bag or V type filter rather than evenly throughout the
media. As air flows into an extended area filter, it must bend
or change direction to go through the sides of the filter.
Because of their inertia many heavy particles resist change in
direction and tend to continue on in their original paths to the
back of the filter. This phenomenon causes the filter
resistance to rise less sharply than would be expected on the
basis of extended area alone.

Loading, Depth The phenomenon by which the contaminants in the air load
throughout the depth of the media, thus giving longer life than
face loading.

Loading, Face The phenomenon by which contaminants in the air load up on

the surface of the filter (sometimes referred to as Blinding),
causing an abnormally rapid rise in resistance.

Manometer Gage used to measure air pressure.

Media Velocity see Velocity, Media.

Media Design see Velocity, Media Design.

Velocity

Micron or Unit of length equal to one millionth of a meter (0.000039-

Micron Size inch). Generally used to classify the size of aerosols.

3/72 27
 Engineering Report: H242

Migration Process by which the adhesive or oily substance releases

itself from the media fibers and enters the air stream and
becomes a contaminant. Migration may cause clogged coils
and dirty ducts as the adhesive collects in the system.

Plenum Chamber see Apparatus House.

Pressure Drop Difference in static air pressure between the upstream and
downstream sides of a filter due to the resistance of the filter.
Usually measured in inches of water.

Resistance Air passing under pressure through an air handling system

must be directed through a number of ducts, the filter bank,
and other obstacles which tend to restrict its flow because of
friction. This restriction of the rate of flow is referred to as
resistance. There is a total resistance related to the entire
system as well as the resistance caused by the filter or filter
bank itself. The general reference to resistance is that
provided by the filter or filter bank, measured in inches of
water.

Safing Sheet metal used to seal off a filter bank when the filter bank
does not run from floor to ceiling or wall to wall in the plenum.

Unloading Process by which dirt, originally stopped by the filter, is

released back into the air stream. Also called blow-off .

Velocity see Feet Per Minute (fpm).

Velocity, The actual velocity of the air as it passes through the filtration
Approach bank area. Approach velocity can be determined by dividing
the cfm of a system by the area of the filter bank opening. A
20,000 cfm system with a filter bank opening 10 feet high by
10 feet wide (an area of 100 sq. ft) would have an approach
velocity of 200 feet per minute as shown below:

20,000 Cubic Feet Per Minute (cfm)

= 200 Feet Per Minute
100 Square Feet

3/72 28
 Engineering Report: H242

Velocity, Media The actual velocity of the air as it passes through one square
foot of media. This can be lower than the approach velocity
where an extended area of V bank system is used. To
determine the media velocity use the following formula:

Total cfm
= Media Velocity
Total Square Feet of Media

Velocity, The media velocity which, through experience, has been

Media Design shown to give the best results in terms of filter life and
efficiency.

Velometer Instrument used to measure the velocity of air.

3/72 29