September 1999
EnvironmentalTechnology
Verification Test Method
General Ventilation Filters
Prepared by
jm
Research Triangle Institute
Under a Cooperative Agreement with
oEPA U.S. Environmental Protection Agency
-------
Notice
This document has been subjected to the U.S. Environmental Protection Agency's quality assurance
and administrative reviews and has been approved for publication.
The EPA, through its Office of Research and Development (ORD), partially funded and managed the
extramural research described here under Cooperative Agreement No. CR 822870.
Mention of trade names or commercial products does not constitute endorsement or recommendation
by EPA or RTI for use.
This document is copyrighted by Research Triangle Institute; however, the Institute agrees to give to the
EPA a non-exclusive, royalty-free, irrevocable license to reproduce and sell the document and to
authorize others to do so, too.
-------
Environmental Technology
Verification Test Method
General Ventilation Filters
Prepared by
Jenni M. Elion
James T. Hanley
David S. Ensor
Deborah L. Franke
Research Triangle Institute
Research Triangle Park, NC 27709-2194
EPA Cooperative Agreement CR 822870
EPA Project Officer: Dr. Leslie Sparks
Air Pollution Prevention and Control Division
National Risk Management Research Laboratory
Research Triangle Park, NC 27711
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page ii of vi
Foreword
The Environmental Technology Verification Test Method, General Ventilation Filters, provides
guidance for verification tests. The test protocol is intended to be a standalone document.
Reference is made in the protocol to the ASHRAE 52.2P "Method of Testing General Ventilation Air-
cleaning Devices for Removal Efficiency by Particle Size" Fourth Composite Working Draft issued in
May 1998. The ASHRAE 52.2P document was issued solely for the purpose of soliciting review
comments and is not a standard. As a result, ASHRAE 52.2P is not currently available from
ASHRAE. However certain test specifications such as particle size distribution parameters and quality
assurance requirements from ASHRAE 52.2P are used in this protocol with attribution and appropriate
caveats. The reason for this approach is that it is believed that the ASHRAE 52.2P standard will likely
not be substantially modified in these fundamental areas before being issued in 1999. It is important to
support our industrial stakeholders by ensuring that the data developed under the ETV program will be
consistent with eventual commercial practice. This approach has been implemented with approval by
ASHRAE.
There are two technical points of exception between this ETV test protocol and the ASHRAE 52.2P
method:
a) The first dust loading step in ASHRAE 52.2P or "conditioning step" has been the subject of review
and research during the last 6 months. In particular, it was found that ASHRAE loading dust
appeared to enhance the performance of certain kinds of media filters over that experienced when
filtering ambient or indoor paniculate matter. As a result the first loading step has been modified to
use a submicrometer aerosol for conditioning.
b) The "Minimum Efficiency Reporting Value" (MERV) is unique to ASHRAE 52.2P and for that
reason it is inappropriate to include this reporting method in the protocol. The protocol will not
include the MERV but information will be available in the verification report to calculate this result if
desired.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page iii of vi
Table of Contents
Foreword ii
1.0 PURPOSE 1
2.0 SCOPE 1
3.0 REFERENCED DOCUMENTS 1
4.0 DEFINITIONS AND ACRONYMS 1
4.1 Definitions 1
4.2 Acronyms 3
5.0 TEST APPARATUS 3
5.1 Introduction 4
5.2 Test Duct 4
5.3 Optical Particle Counter 6
5.4 Challenge Aerosol Generation 6
5.5 Aerosol Sampling System 7
5.6 Conditioning Aerosol Generation 8
5.7 Dust Loading System 9
5.7.1 Loading Dust 9
5.7.2 Dust Loading Apparatus 9
6.0 SYSTEM QUALIFICATION TESTING 9
6.1 Airflow Uniformity in the Test Duct 10
6.2 Aerosol Concentration Uniformity in the Test Duct 11
6.3 Downstream Mixing of Aerosol 11
6.4 Aerosol Generator Response Time 12
6.5 Upper Concentration Limit of the OPC 12
6.6 100 % Efficiency Test 13
6.7 Correlation Ratio Test 13
© Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page iv of vi
6.8 Test Duct Air Leakage Test 13
7.0 TEST METHOD 13
7.1 Qualification Testing 13
7.1.1 100 %Efficiency Test 14
7.1.2 OPC Sizing Accuracy Check 14
7.1.3 OPC Zero Count 15
7.1.4 Background Particle Count 15
7.1.5 Correlation Ratio Test 15
7.1.6 Pressure Drop 15
7.2 Test Procedure 15
8.0 AEROSOL CONDITIONING 17
9.0 DUST LOADING 17
10.0 CALCULATIONS 17
10.1 Nomenclature 17
10.2 Equations 17
10.3 Acceptance Criteria for Data 18
11.0 DOCUMENTATION AND REPORTING 19
11.1 Raw Data 19
11.2 Calculated Data 19
11.3 Reported Data 19
12.0 BIBLIOGRAPHY 20
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page v of vi
List of Figures
1. Schematic illustration of test apparatus 5
2. Details of the example aerosol generating system 7
3. Schematic diagram of the example aerosol sampling system 8
4. Nine-point equal area sampling grid for airflow and aerosol uniformity assessment 11
5. Nine-point perimeter injection grid to assess downstream mixing 12
List of Tables
1. Minimum Sizing Channels Required for OPC 6
2. System Qualification Test Requirements 10
3. Ongoing Qualification Test Requirements 14
4. Testing Sequence 16
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page vi of vi
This page is left intentionally blank.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 1 of 20
1.0 PURPOSE
The test method establishes a procedure for evaluating the efficiency of general ventilation filters as a
function of particle size and as a function of loading.
2.0 SCOPE
The method provides a procedure for counting airborne particles from 0.3 jim to 10 jim in diameter
both upstream and downstream of an air cleaning device and calculating the removal efficiency by
particle size.
The method provides a procedure for determining efficiency of new unused air cleaning devices and of
the same devices after subsequent particle loading.
3.0 REFERENCED DOCUMENTS
ANSI/ASMEN510, Testing of'Nuclear Air-Cleaning Systems. American Society of Mechanical
Engineers, New York, NY, 1980.
ASHRAE Standard 52.1, Gravimetric and Dust-Spot Procedures for Testing Air-Cleaning
Devices Used in General Ventilation for Removing P articulate Matter. American Society of
Heating, Refrigerating and Air-Conditioning Engineers, Inc., Atlanta, GA, 1992.
ASHRAE Proposed Standard 52.2P, Method of Testing General Ventilation Air-Cleaning
Devices for Removal Efficiency by Particle Size, Fourth Composite Working Draft. American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., Atlanta, GA, 1998. (Draft
released for a limited time for review and is currently not available from ASHRAE.)
IEST-RP-CC007.1. Testing ULPA Filters. Institute of Environmental Sciences and Technology,
Mount Prospect, IL, 1992.
4.0 DEFINITIONS AND ACRONYMS
4.1 Definitions
Some terms are defined below for the purposes of this method. When definitions are not provided,
common usage shall apply.
airflow rate: the actual volume of test air passing through the device per unit of time, expressed in
m3/s [ftYmin (cfm)], to three significant figures.
charge neutralizer: a device that brings the charge distribution of the aerosol to a Boltzman charge
distribution. This represents the charge distribution of the ambient aerosol.
coefficient of variation: standard deviation of a group of measurements divided by the mean,
expressed as a percentage.
conditioning aerosol: the generated aerosol used during the conditioning step. The aerosol particles
are composed of potassium chloride having a mean diameter of approximately 0.1 jam and generated
for a duration of 8 hours at a rate that produces a concentration of about 200,000 particles/cm3 in the
challenge air stream.
conditioning step: loading procedure performed after the initial efficiency test to reveal changes in
efficiency that a filter may undergo during the early stages of use.
© Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 2 of 20
correlation ratio: the ratio of downstream to upstream particle counts without the test device installed
in the test duct and determined from the average of at least three samples. This ratio is used to correct
for any bias between upstream and downstream sampling and counting systems.
correlation ratio data acceptance criteria: criteria used to determine the adequacy of the
correlation data.
device: throughout this method the word "device" means air-cleaning equipment used in general
ventilation for the removal of particles, specifically, the air cleaner being tested.
disposable air filters: filters that are designed to operate through a specified performance range and
then be discarded and replaced.
dust increment: the amount of dust fed during the loading procedure.
face area: the gross area of the device exposed to airflow. This area is measured in a plane
perpendicular to the axis of the test duct or the specified direction of airflow approaching the device.
All internal flanges are a part of this area, but items such as mounting hardware and electrical raceways
normally mounted out of the airstream are not included. Face area is measured in m2 (ft2), to three
significant figures.
face velocity: the rate of air movement at the face of the device (airflow rate divided by face area)
expressed in m/s (fpm), to three significant figures.
final filter: a filter used to collect the loading dust that has passed though a device during the test
procedure.
final resistance: the predetermined resistance to airflow of the air cleaning device at which the test is
terminated and results calculated, expressed in Pa (in. H2O).
general ventilation: the process of moving air into or about a space or removing it from the space.
The source of ventilation air is either air from outside the space, or recirculated air, or a combination of
these.
initial resistance: the pressure loss of the device operating at a specified airflow rate with no dust
load, expressed in Pa (in. of H2O).
isokinetic sampling: sampling in which the flow in the sampler inlet is at the same velocity as the flow
being sampled.
loading dust: a compounded synthetic dust used for air cleaner loading.
media: for a fibrous-type air cleaner, media is that part of the device that is the actual dust-removing
agent. Webs of spun fiberglass and papers are examples of air filter media.
media velocity: the speed at which air moves through the filter media (airflow rate divided by net
effective filtering area). The term is not applicable to plate-type electronic air cleaners. Media velocity
is measured in m/s (fpm), to three significant figures.
net effective filtering area: the total area in the device on which the dust collects. For devices using
fibrous media, it is the net upstream area of the media exposed to airflow measured in the plane or
general surface of the media. Net effective area excludes the area blocked by sealants, flanges, or
supports. In electronic air cleaners, it is the total exposed surface of those electrodes available for dust
© Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 3 of 20
precipitation, including the ionizing section but excluding the supports, holes, and insulators. Net
effective filtering area is measured in m2 (ft2), to three significant figures.
particle size: the polystyrene latex (PSL) light scattering equivalent diameter expressed as a diameter
in micrometers, (|im, 10"6 meters).
penetration: the fraction (percentage) of particles that pass through the air cleaner.
penetration data acceptance criteria: criteria used to determine the adequacy of the penetration
data.
polydisperse: an aerosol characterized by a wide distribution of particle sizes.
rated airflow: the airflow rate in m3/s (cfm) at which the device is tested.
rated final resistance: the operating pressure loss at the specified airflow rate at which a device
should be replaced or renewed, as recommended by the manufacturer, expressed in Pa (in. H2O).
reference filters: dry media-type filters that are carefully measured for resistance and initial efficiency
immediately after a test system is qualified. These filters serve as references to insure that the test
system continues to operate as it did when it was qualified.
resistance: the loss of static pressure caused by the device operating at a stated airflow rate,
expressed in Pascals (in. H2O) to an accuracy of ±2.5%
test aerosol: polydisperse solid-phase (i.e., dry) potassium chloride (KC1) particles generated from
an aqueous solution, used in this method to determine the particle size efficiency of the device under
test.
test rig: the total assembly consisting of the test duct, the aerosol generator, the loading dust feeder,
the particle counter(s) and associated accessories, instrumentation, and monitoring equipment.
4.2 Acronyms
ANSI American National Standards Institute
ASHRAE American Society of Heating, Refrigerating and Air-Conditioning Engineers
ASME American Society of Mechanical Engineers
ASTM American Society for Testing and Materials
CV coefficient of variance
HEPA high efficiency particulate air
LEST Institute of Environmental Sciences and Technology
OPC optical particle counter
PSL polystyrene latex (spheres)
RTI Research Triangle Institute
ULP A ultra low penetration air
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 4 of 20
5.0 TEST APPARATUS
5.1 Introduction
The test rig combines the aerosol generation, sampling, and overall duct configuration developed by
RTI as part of the ASHRAE- and EPA-sponsored research with a high airflow capacity (3000 cfm)
blower, mixing baffles, and airflow measurement device of the large-scale test duct used in ASHRAE
Standard 52.1. Features of the test duct (Figure 1) include:
• Positive pressure to minimize room air infiltration.
• Inlet air drawn from indoors or recirculated air to maintain temperature and humidity within desired
range.
• HEPA-filtered inlet to remove ambient aerosol.
• HEPA-filtered exhaust to allow indoor discharge.
• Artificially generated, solid-phase, polydisperse potassium chloride salt challenge aerosol.
• Low particle loss sampling system consisting of a single optical particle counter (OPC) or dual
OPCs.
• Inclusion of a downstream mixing baffle to ensure well-mixed aerosol conditions at the downstream
sample probes.
• 180° bend in downstream duct to bring upstream and downstream sample locations in close
proximity to each other, greatly reducing sample line length and sample losses and facilitating use of
a single OPC.
5.2 Test Duct
The following is an example of a test rig successfully used for filter testing. The test rig's ducting is a
square cross-section 610 x 610 mm (24 x 24 in.), 14 gauge stainless steel. The upstream and
downstream mixers are per ASHRAE 52.1. The blower has a flow capacity of 1420 L/s at 3.2 kPa
(3000 cfm at 13 in. H2O). The inlet and outlet filter banks consist of two 610 x 610 x 51 mm (24 x 24
x 2 in.) prefilters and two 610 x 610 x 305 mm (24 x 24 x 12 in.) high efficiency particulate air (HEPA)
filters rated at 940 L/s (2000 cfm) each. Airflow is measured with an American Society of Mechanical
Engineers (ASME) flow orifice as specified in ASHRAE 52.1. The system includes a means of
controlling airflow. Room or recirculated air is used as the source, and the flow is exhausted back into
the room after passing though the exhaust HEPA bank. Air may be recirculated directly without
exhausting into the room if appropriate cooling or air conditioning is employed.
To mix the test aerosol with the air stream, an orifice plate and mixing baffle per ASHRAE 52.1 should
be used immediately downstream of the aerosol injection point. An identical orifice plate/mixing baffle
should be added after the 180° bend or at the same relative point in a straight-line system.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters
Page 5 of 20
(2,OOQCFMHEFAB
WITH PREFILTERS)
OUTLET FILTER BANK
TOP VIEW
OPTICAL
PARTICLE
COUNTER
ROOM
AIR
EXHAUST
TO ROOM
ASME
NOZZLE
DOWNSTREAM
MIXER
\
i, '_._' I n 5
t^=T x >
INLET
i 0 ji ^ -I | ' ^
A -
1
FILTER
BANK
(2-2,000 CFM HE PAS
2 ' WITH PREFILTERS}
FLOW CONTROL
VALVE
\ \ DEVICE BACKUP
\ SECTION FILTER
\ HOLDER
\ (USED WHEN
^y DUST-LOADING)
UPSTREAM MIXER
AEROSOL INJECTION
LIQUID SALT
SOLUTION
Q^X
SPRAY NOZZLE gpg V|EW
BB^. T ^^— _ _ . .^^^^^^ . . ^
^ •* ^i^ ^" CO MFES S ED Al R
CHARGE
NEUTRALIZER \^. r
—
/^^— s^\
( J6H
Y^y
t
,/ |5
I >^ Ix
x |
J
BLOWER
(3,000 CFM
" "
"={=> ,i ^ g
^S >
INLET FILTER
BANK
(2-2,000 CFM HEPAg
W ITH PREFILTERS}
FLOW CONTROL
VALVE
\ DEVICE BACKUP
!
\ SECTION FILTER ' =
\ HOLDER 180°
(USED WHEN LOOP
UPSTREAM MIXER DUST LOADING)
AEROSOL INJECTIOM
11690-Elim-lestriB
Figure 1. Schematic illustration of test apparatus
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters
Page 6 of 20
5.3 Optical Particle Counter
Aerosol concentration should be measured with a Climet Instruments Model 226 OPC or equivalent.
This OPC uses a high intensity illumination source and has a wide collection angle for the scattered light.
The OPC's sampling rate should be 0.12 L/s (0.25 cfm).
The output of the OPC should input to a Climet Instruments Model 8040 multichannel analyzer or
equivalent capable of handling a minimum of 12 sizing channels. The particle diameter range boundaries
required are listed in Table 1.
Table 1. Minimum Sizing Channels Required for OPCa
Range
1
2
3
4
5
6
7
8
9
10
11
12
Diameter Range (um)
Lower Limit
0.30
0.40
0.55
0.70
1.00
1.30
1.60
2.20
3.00
4.00
5.50
7.00
Upper Limit
0.40
0.55
0.70
1.00
1.30
1.60
2.20
3.00
4.00
5.50
7.00
10.00
Geometric Mean
Particle Diameter
(urn)
0.35
0.47
0.62
0.84
1.14
1.44
1.88
2.57
3.46
4.69
6.20
8.37
Trom ASHRAE 52.2P (1998)
5.4 Challenge Aerosol Generation
The challenge aerosol should be generated by a TSI generator or equivalent, as illustrated in Figure 2.
The test aerosol is solid-phase dry potassium chloride (KC1) generated from an aqueous solution. The
aerosol is generated by nebulizing an aqueous KC1 solution with a two fluid (air and water) air atomizing
nozzle (Spray Systems 1A J siphon spray nozzle or equivalent). The hole in the air cap is increased to
3.2 mm (0.125 in.) inner diameter.
The nozzle is positioned at the top of a 0.30 m (12 in.) diameter, 1.3 m (51 in.) tall transparent acrylic
spray tower through which drying air flows at 1.89 L/s (4 cfm). After generation, the aerosol passes
through a TSI, Inc., Model 3054 aerosol neutralizer or equivalent to neutralize any electrostatic charge
on the aerosol. To improve the mixing of the aerosol with the air stream, the aerosol should be injected
counter to the airflow as illustrated in Figure 2.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters
Page 7 of 20
The KC1 solution is prepared by combining approximately 300 g KC1 with 1 L deionized water. The
solution is fed to the atomizing nozzle by means of a metering pump. Varying the operating air pressure
of the generator allows control of the mean diameter of the challenge aerosol.
Spray Tower
30 cm Diam 13D cm Tall
(12 in. Diam 51 in. Tall)
Syringe Pump
Air Atomizing Nozzle
AaroEol Charge
Neutralize '
\
iH
*®
^7
A
r
< Atom king Air 0. 5 L/s
Dryjng (1 CFM) Nominal
Air
J'^oui > 7 HEPA Filter Ca?6"168
(4 UrlV)J/ . . /
~~^ i
I
M r
©
0
« ^
© ®
A r Control Panel
^^^ Clean Dry Com
ftir boura
3.8 cm ID
(1.5 in.)
15.2 cm OD Disk
Test Duct
HEPA
Filter
Bank
Figure 2. Details of the example aerosol generating system
5.5 Aerosol Sampling System
Losses in the aerosol sampling system shall not exceed 50%. The example shown is constructed of 14
mm (0.55 in.) inner diameter stainless steel lines or equivalent, using gradual bends [radius of curvature
= 57 mm (2.25 in.)] when needed. The "Y" fitting connecting the upstream and downstream lines to
the OPC merges gradually to minimize particle loss in the intersection of the "Y" due to centrifugal or
impaction forces. The fitting may be custom-made per Figure 3.
Immediately above the "Y," an electrically actuated ball valve is installed in each branch (Parker Model
EA Electro-Mechanical Valve Actuator or equivalent). The opening and closing of the valves are
automatically controlled by the OPC's sequential sampling interface board. The valves take
approximately 2 seconds to complete an opening or closing maneuver.
Isokinetic sampling nozzles of the appropriate entrance diameter are placed on the ends of the sample
probes to maintain isokinetic sampling for all the test flow rates.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters
Page 8 of 20
610mm
(24 in.)
/
/
r
stst
Valve
Aclualo
1
610mm
(24 in.)
-»/«-
610mm
(24 in.)
Downstream Duct
Duct
Wall^
pie Lines
\
/ Ball
S Valve
•
7
f
>
t>
/
/
610mm
(24 in.)
Upstream Duct
Machined "Y"
Data Acquisition
Computer
Figure 3. Schematic diagram of the example aerosol sampling system
5.6 Conditioning Aerosol Generation
The conditioning aerosol should be generated by a 3-nozzle (12-jet) Laskin nozzle or equivalent. The
test aerosol is solid-phase dry potassium chloride (KC1) generated from a 1% aqueous solution. The
KC1 solution is prepared by combining KC1 and deionized water at the ratio of 10 g KC1 to 1 L
deionized water.
The output of the Laskin nozzle should be injected into the spray tower (as shown earlier in Figure 2)
into which drying air flows at 1.89 L/s (4 cfm). After generation, the aerosol passes through a TSI,
Inc., Model 3054 aerosol neutralizer or equivalent to neutralize any electrostatic charge on the aerosol.
To improve the mixing of the aerosol with the air stream, the aerosol should be injected counter to the
airflow as was illustrated in Figure 2.
The operating air pressure of the Laskin nozzle is adjusted to provide a particle concentration in the
challenge air stream of 125,000 particles/cm3 ± 20%. This challenge condition is maintained for 8
hours. Means of replenishing the aqueous solution within the Laskin nozzle should be used to allow for
continuous operation over the entire 8-hour period.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 9 of 20
5.7 Dust Loading System
5.7.1 Loading Dust
Loading dust can be obtained commercially under the trade name ASHRAE® dust.
5.7.2 Dust Loading Apparatus
The dust loading apparatus is described in ASHRAE Standard 52.1.
6.0 SYSTEM QUALIFICATION TESTING
Qualification testing shall be completed before filter testing is attempted. The purpose of the system
qualification tests is to quantify that the test rig and sampling procedures are capable of providing
reliable fractional penetration measurements. The qualification tests are based on those in IEST-RP-
CC7.1.
Qualification tests are performed for the following:
• Air flow uniformity in the test duct,
• Aerosol concentration uniformity in the test duct,
• Downstream detection of the aerosol,
• Aerosol generator response time,
• Upper concentration limit of the OPC,
100% efficiency test,
Correlation ratio test, and
• Duct leak test.
The requirements for system qualification, summarized in Table 2, are as specified in ASHRAE 52.2P,
a draft review standard that may be changed before approval.
These tests must be repeated after any change that may alter system performance.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters
Page 10 of 20
Table 2. System Qualification Test Requirements
Parameter
Air Flow Uniformity:
Based on traverse measurements made over a 9-pt. equal-area grid at each test flow rate.
Aerosol Concentration Uniformity:
Based on traverse measurements made over a 9-pt. equal-area grid at each test flow rate.
Downstream Mixing:
Based on a 9-pt. perimeter injection grid and cumulative downstream particle count.
Aerosol Generator Response Time
Upper Concentration Limit:
Based on limiting the concentration to below the concentration corresponding to the
onset of coincidence error in the OPC.
100% Efficiency:Based on HEPA filter test
Correlation Ratio:
Based on five replicate tests at each flow rate.
Duct Leakage:
Ratio of leak rate to test airflow rate.
Level Required1
CV<10%
CV<15%
CV<10%
No pre -determined level.
No pre -determined level.
>99%
for all particle sizes
0.3 to 1.0 ^m; 0.90-1. 10
1.0 to 3.0^; 0.80-1.20
3.0 to 10 urn; 0.70-1.30
<1.0%
1 These levels conform to the levels required in ASHRAE 52.2P.
6.1 Airflow Uniformity in the Test Duct
The uniformity of the challenge airflow is determined by a nine-point traverse (Figure 4) immediately
upstream of the device section. Airflow is measured with an Alnor Model 8500D-II Thermo-
anemometer or equivalent. At each grid point, the average of 10 air flow readings taken over a 1-
minute period is recorded. After measuring the 1-min average at each gridpoint, the traverse is
repeated two more times, providing triplicate 1-min averages at each point. The triplicate values are
then averaged for each point. This is done for airflows of 470, 940, and 1420 L/s (1000, 2000, and
3000 cfm).
For all three flow rates, the velocity at each gridpoint shall have a coefficient of variance (CV) value less
than 10 %.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters
Page 11 of 20
610 mm
(24 in)
610mm
(24 in)
•
4
•
2
8
•
6
Figure 4. Nine-point equal area sampling grid for airflow and aerosol uniformity assessment
6.2 Aerosol Concentration Uniformity in the Test Duct
The uniformity of the challenge aerosol concentration across the duct cross section is determined by a
nine-point traverse (same grid points as shown in Figure 4). The aerosol concentration is measured
with the OPC. At each gridpoint, a 1-min sample is taken. After sampling from all nine points, the
traverse is repeated four more times providing a total of five samples from each point. The five values
for each point are then averaged. This is done for airflows of 470, 940, and 1420 L/s (1000, 2000,
and 3000 cfm).
For each flow rate, the CVs for aerosol concentration uniformity for all particle sizes shall be less than
15%.
6.3 Downstream Mixing of Aerosol
The test is conducted by first installing a FffiPA filter in the device section. Aerosol is then injected
immediately downstream of the FLEPA filter at preselected injection points located around the perimeter
of the test duct and at the center of the duct (Figure 5). Unlike the aerosol traverse, in this test the point
of aerosol injection is traversed and the downstream sampling probe remains stationary in its normal
center-of-duct sampling location. Unlike the grid points for the flow and aerosol traverses, this grid
pattern includes points very close to the test duct walls. A Collison nebulizer generates the aerosol and
the OPC measures total aerosol concentration (>0.3 |im) rather than examining the concentration in
each of the OPC's required 12 channels.
Regardless of where the aerosol was injected, the CVs for all downstream readings shall be less than
10%.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters
Page 12 of 20
610mm
(24 in)
610mm
(24 in)
8
*
Figure 5. Nine-point perimeter injection grid to assess downstream mixing.
Perimeter points are 25 mm (1 in.) from duct wall.
6.4 Aerosol Generator Response Time
To ensure that sufficient time is allowed for the aerosol concentration to stabilize during the fractional
penetration tests, the stabilization time for the aerosol to go from the background level to the test level is
measured. This is done with the OPC sampling from the upstream probe.
Similarly, the time to return to the background levels after turning off the generator is measured. When
turning on the generator, the air and liquid to the spray nozzle are shut off. The system may be purged
with drying air and then allowed to run after the drying air has been turned off.
6.5 Upper Concentration Limit of the OPC
A series of initial efficiency tests shall be performed over a range of challenge aerosol concentrations to
determine a total level for the efficiency tests that does not overload the OPC. The lowest total
concentration shall be less than 1% of the instrument's stated total concentration limit. The filters
selected for this test shall have an initial efficiency in the range of 30-70% as measured by the 0.3-0.4
|im diameter size range and greater than 90% efficiency for the 7.0-10.0 jim diameter size range. The
aerosol for these tests shall be generated using the same system and procedures as described in Section
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 13 of 20
5.4. The tests shall be performed over a sufficient range of total challenge concentrations to
demonstrate that the OPC is not overloaded at the intended test concentration.
6.6 100 % Efficiency Test
A 100 % efficiency test is performed as a normal initial efficiency test except that a HEPA filter is used.
The efficiency test consists of the five alternating upstream and downstream background counts (KC1
generator off), ten alternating upstream and downstream sample counts (KC1 generator ON), and five
alternating upstream and downstream background counts (KC1 generator off). Penetration shall not
exceed 1% for any particle size range.
6.7 Correlation Ratio Test
A zero percent efficiency (or blank) test is performed as a normal penetration test except that no air
filter is used. Fifteen 100 % penetration tests are conducted, five at each of three flow rates: 470,
940, and 1420 L/s (1000, 2000, and 3000 cfm). The ratio of downstream to upstream particle counts
for each size range shall fall within the limits specified in Table 2.
6.8 Test Duct Air Leakage Test
The leak rate of the test duct is evaluated by a method similar to American National Standards Institute
(ANSI)/ASME N510. The test duct is sealed immediately upstream of the aerosol injection location
and immediately upstream of the exhaust filter bank by bolting a gasketed solid plate to the duct
opening. Air is then metered into the test duct until the lower test pressure is achieved. The airflow rate
required to maintain constant pressure is measured and taken as the leak rate. This is repeated for the
upper pressure level.
[CAUTION: Excessively high pressures may explode the test rig.]
For the airflow measurement to be representative of the airflow through the air cleaner, air leakage from
the test duct should not exceed 1 % of the total flow.
7.0 TEST METHOD
7.1 Qualification Testing
Additional qualification testing shall be performed on a continuing basis. The purpose of these ongoing
tests is to quantify that the test rig and sampling procedures are providing reliable fractional penetration
measurements. The qualification tests are based on those in IEST-RP-CC7.1.
Qualification tests are performed for the following:
• 100 % efficiency,
OPC sizing accuracy,
OPC zero count,
• Background particle count,
• Correlation ratio test, and
• Pressure drop across empty test section.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters
Page 14 of 20
The frequency and requirements for the ongoing qualification tests are summarized in Table 3.
Table 3. Ongoing Qualification Test Requirements
Parameter
100% Efficiency:
Based on HEP A filter test
OPC sizing accuracy check:
Sample aerosolized PSL spheres.
OPC zero count:
OPC samples HEPA-filtered air.
Correlation ratio:
Pressure drop:
Across empty test section
Upstream background particle counts:
KC1 aerosol generator turned off.
Downstream background particle counts:
KC1 aerosol generator turned off.
Frequency1
Monthly
Daily
Each test
Each test
Each test
Each test
Each test
Control Limits3
>99%
for all particle sizes
Relative maximum must appear in
appropriate sizing channel
<10 counts/minute over 0.30 to 10 (im range
0.3 to 1.0 ^m; 0.90-1. 10
1.0 to 3.0 ^m; 0.80-1.20
3.0 to 10 urn; 0.70-1.30
<8 Pa (0.03 in. H2O)
<5% of upstream challenge counts
in each channef
<5% of upstream challenge counts
in each channef
a These levels conform to the levels required in ASHRAE 52.2P.
b For the initial efficiency tests only, both the upstream and downstream
than 5% of the upstream challenge particle counts in each channel when
requirement does not apply after conditioning or incremental dust loadin
background particle counts must be less
the aerosol generator is on. This
7.1.1 100 % Efficiency Test
A 100 % efficiency test is performed as a normal penetration test except that a HEPA filter is used.
Cross contamination between the upstream and downstream samples and insufficient purging between
samples are potential conditions that could limit the ability to meet this requirement.
On at least a monthly basis, a HEPA filter will be tested following the same procedures as used for
testing ventilation filters. Penetration shall not exceed 1% for any particle size range.
7.1.2 OPC Sizing Accuracy Check
The sizing accuracy of the OPC is checked by sampling an aerosol containing monodispersed
polystyrene latex (PSL) spheres of known size.
This is not a calibration but simply a calibration check of the OPC that is performed daily. A relative
maximum particle count shall appear in the particle counter sizing channel that encompasses the PSL
diameter.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 15 of 20
7.1.3 OPC Zero Count
The OPC's zero count level will be measured by sampling HEPA-filtered air. This will be done by
attaching a HEPA capsule directly to the OPC sample inlet and/or by sampling the HEPA-filtered air of
the test duct when the aerosol generator is turned off.
This check is performed at the start of each efficiency test. The counts will be summed to confirm that
they are below 10 per minute. On the test run sheet, the operator denotes if an acceptable zero count
was achieved.
7.1.4 Background Particle Count
Background counts will be made before and after generating test aerosols during every efficiency test.
Upstream and downstream sampling is done sequentially, starting with an upstream sample Ub followed
by a downstream sample Dh alternating back and forth five times.
7.1.5 Correlation Ratio Test
A zero percent efficiency (or blank) test is performed as a normal penetration test except that no air
filter is used. It is performed immediately prior to each filter test. The purpose is to ensure that the bias
between the upstream and downstream samples is within acceptable limits. The ratio of downstream to
upstream particle counts for each size range shall fall within the limits specified in Table 3. If they are
found to be out of range, the data will be inspected for possible clues, and corrective action (such as
cleaning the sample lines) may be required to obtain acceptable results. The filter test cannot be
conducted until an acceptable 100% test is achieved.
7.1.6 Pressure Drop
The pressure drop across the empty test section shall be measured as part of each correlation ratio test.
The measured pressure drop across the empty test section shall be less than 8 Pa (0.03 in. H2O).
7.2 Test Procedure
1. Warm up OPC and install proper sample tips for isokinetic sampling.
2. Perform monthly 100% efficiency test (if necessary) to confirm that it is within the range established
in the qualification tests. If not, take corrective action and repeat this step.
3. Perform daily calibration check of OPC sizing accuracy to confirm that it is within the range
established in the qualification tests. If not, take corrective action and repeat this step.
4. Perform OPC zero check to confirm that it is within the range established in the qualification tests.
If not, take corrective action and repeat this step.
5. Perform correlation ratio test and record pressure drop to confirm that they are within the range
established in the qualification tests. If not, take corrective action and repeat this step.
6. Install an air-cleaner test device, and bring the test duct to the desired flow rate.
7. Verify that the test duct's relative humidity and temperature are within their respective operating
ranges. Run the test duct for 15 minutes for the air cleaner to equilibrate.
8. With the aerosol generator off, obtain five measurements of the upstream and downstream
background particle counts. For some dust-loaded air cleaners, shedding of the loading dust may
© Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters
Page 16 of 20
lead to background counts in some sizing channels that are known (from experience) to be too high
(e.g, exceed 10% of the expected downstream concentration when the aerosol generator is turned
on). In this case, repeating step 6 may bring the background counts down to acceptable levels.
9. Turn on the aerosol generator and allow it to run for 10 minutes to stabilize.
10. After the stabilization period, obtain 10 upstream and 10 downstream particle counts. These are
obtained in an up-down-up-down fashion until 10 of each are obtained. Each sample shall last 1
minute.
11. Turn off the aerosol generator. Wait 10 minutes, then obtain five additional upstream and
downstream background measurements.
12. Verify that data meet acceptance criteria for minimum upstream particle counts and for penetration.
The acceptance criteria are outlined in Section 10.3.
After the initial efficiency is tested, the filter is conditioned and its efficiency is tested again. The filter is
then subjected to four incremental dust loadings; the filtration efficiency is measured after each loading.
The final dust loading step is completed when the filter's resistance to airflow reaches a predetermined
value specified by the manufacturer. The complete test sequence is illustrated in Table 4.
Table 4. Testing Sequence
Step: Description
1: Correlation ratio
2: Resistance vs. airflow
3: Initial Efficiency"
4: Conditioning
5: Efficiency after Conditioning3
6: First dust loadingb
7: Efficiency after first dust loading3
8: Second dust loadingb
9: Efficiency after second dust loading3
10: Third dust loadingb
1 1 : Efficiency after third dust loading3
12: Fourth dust loadingb
13: Efficiency after fourth dust loading3
Air
Cleaner
None
Installed
Installed
Installed
Installed
Installed
Installed
Installed
Installed
Installed
Installed
Installed
Installed
Conditioning
Generator
Off
Off
Off
ON
Off
Off
Off
Off
Off
Off
Off
Off
Off
KC1
Generator
ON
Off
ON
Off
ON
Off
ON
Off
ON
Off
ON
Off
ON
Dust
Feeder
Off
Off
Off
Off
Off
ON
Off
ON
Off
ON
Off
ON
Off
a Each efficiency test consists of the background counts (KC1 generator off), sample counts (KC1 generator ON), and
background counts (KC1 generator off).
b Prior to each dust loading, the duct airflow is turned off, the final filter is installed, and the particle counter inlet
probes are capped. The duct airflow is then resumed. After each dust loading, the duct airflow is turned off, the
particle counter inlet probes are uncapped, and the final filter is removed. The duct airflow is then resumed.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 17 of 20
8.0 AEROSOL CONDITIONING
After being tested for initial efficiency, the filter shall be exposed to the conditioning aerosol as
described in Section 5.6. This corresponds with Step 4 in Table 4.
After aerosol conditioning, the fractional efficiency is measured as described in Section 7.2.
9.0 DUST LOADING
The procedure for dust loading is described in ASHRAE Standard 52.1. Table 4 indicates that dust
loading occurs as Steps 6, 8, 10, and 12.
After each dust increment, the fractional efficiency is measured as described in Section 7.2.
10.0 CALCULATIONS
10.1 Nomenclature
P0 = Observed penetration
D = Downstream particle count
Db = Downstream background count
U = Upstream particle count
Ub = Upstream background count
P = Penetration corrected for P100 value
P100 = 100% penetration value determined in system qualification tests
<7 = Sample standard deviation
CV = Coefficient of variance
CVU = Upstream coefficient of variance
CVd = Downstream coefficient of variance
Overbar "~" denotes arithmetic mean of quantity
10.2 Equations
Analysis of each test involves the following quantities:
• P100 value for each sizing channel from the system qualification tests,
• 10 upstream background values,
10 downstream background values,
10 upstream values with the aerosol generator on, and
• 10 downstream values with the aerosol generator on.
Using the values associated with each sizing channel, the observed penetration associated with each
particle sizing channel is calculated as shown in Eq. (1). The corrected penetration for each particle
sizing channel is calculated as shown in Eq. (2).
© Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters _ Page 1 8 of 20
(2) P=P-
"lOO
Most often, the background levels will be small compared to the values when the aerosol generator is
on. However, for some types of air cleaners, shedding of the loading dust may result in high
background levels over portions of the size range. The filtration efficiency is calculated as shown in Eq.
(3).
(3) Filtration Efficiency (%) = (l - P) x 100
Negative filtration efficiencies are reported as zero.
10.3 Acceptance Criteria for Data
After each filtration efficiency test, data must be checked against the following criteria before
proceeding. If the acceptance criteria are not satisfied, the test must be repeated. Repeating the test
before proceeding will not compromise the integrity of the method.
For each channel, a minimum total of 500 particle counts for the challenge aerosol is required. The
total is summed over the 10 upstream readings taken during testing, as shown in Eq. (4).
10
(4) For each channel, ^U, > 500 where i = sample reading
i=\
For each channel of each test, the standard deviation of observed penetration is calculated based on the
CV of the upstream and downstream particle counts and must satisfy the limits shown in Eqs. (5) and
(6).
(5) aPa = P0x CVd2 + CVU2 < 0. 1 for 0.3 -3.0^m diameter
(6) ap = P0 x -cVd + CVU < 0. 3 for > 3 . 0 \m diameter
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 19 of 20
11.0 DOCUMENTATION AND REPORTING
The test report will consist of calculated and reported data. Raw data shall be included as an appendix.
11.1 Raw Data
Test data and notes are recorded in laboratory notebooks. Test personnel shall record:
• Testing date and location
• Manufacturer and model number
• Physical description of filter
• Pressure drop across the filter
• Qualification tests
• Ancillary environmental data
Raw data recorded on instrument printouts are taped into the laboratory notebook. These include:
• Upstream particle count
• Downstream particle count
• Upstream particle count, background
• Downstream particle count, background
11.2 Calculated Data
Calculated data shall include:
observed penetration,
• corrected penetration,
• sample standard deviation, and
coefficient of variation for all particle diameters.
11.3 Reported Data
Reported data shall include:
• the testing date and location,
a physical description of the device,
• filtration efficiency curves from each test and their averages,
• tabulated efficiency data,
• efficiencies at particle diameters,
• results of control tests, and
• the pressure drop across the filter.
1 Research Triangle Institute - 1999
-------
Research Triangle Institute
ET V Test Method
General Ventilation Filters Page 20 of 20
12.0 BIBLIOGRAPHY
ANSI/ASMEN510, Testing of'Nuclear Air-Cleaning Systems. American Society of Mechanical
Engineers, New York, NY, 1980.
ASHRAE Fundamentals Handbook. American Society of Heating, Refrigerating and Air-
Conditioning Engineers, Inc., Atlanta, GA, 1989.
ASHRAE Standard 52.1, Gravimetric and Dust-Spot Procedures for Testing Air-Cleaning
Devices Used in General Ventilation for Removing P articulate Matter. American Society of
Heating, Refrigerating and Air-Conditioning Engineers, Inc., Atlanta, GA, 1992.
ASHRAE Proposed Standard 52.2P, Method of Testing General Ventilation Air-Cleaning
Devices for Removal Efficiency by Particle Size, Fourth Composite Working Draft. American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., Atlanta, GA, 1998. (Draft
released for a limited time for review and is currently not available from ASHRAE.)
Hanley, J. T., D. D. Smith, P. A. Lawless, D. S. Ensor, and L. E. Sparks. "A Fundamental Evaluation
of an Electronic Air Cleaner." In: Proceedings of the Fifth International Conference on Indoor Air
Quality and Climate. Vol. 3, pp. 145-150, 1990.
IEST-RP-CC007.1. Testing ULPA Filters. Institute of Environmental Sciences and Technology,
Mount Prospect, IL, 1992.
Owen, M. K., D. S. Ensor, and L. E. Sparks. "Airborne Particle Sizes and Sources Found in Indoor
Air." Atmospheric Environment 2<14(12):2149-2162, 1992.
Research Triangle Institute. Project Work and Quality Assurance Plan for Fractional Efficiency of
Air Cleaners., EPA Cooperative Agreement CR-817083. Prepared for U.S. EPA, AEERL, Research
Triangle Park, NC, 1993.
Viner, A. S., K. Ramanathan, J. T. Hanley, D. D. Smith, D. S. Ensor, and L. E. Sparks. "Air Cleaners
for Indoor Air Pollution Control." Indoor Air Pollution; Radon, Bioaerosols, & VOCs. J. G. Kay,
G E. Keller, and J. F. Miller, eds., Lewis Publishers, 1991.
1 Research Triangle Institute - 1999
------- |