Joint Committee Issue Paper

NSF Standard(s) Impacted:

Background:
Provide a brief background statement indicating the cause and nature of concern, the impacts identified
relevant to public health, public understanding, etc, and any other reason why the issue should be
considered by the Committee. Reference as appropriate any specific section(s) of the standard(s) that are
related to the issue.
Currently NSF/ANSI 49 only allows HEPA/ULPA filter leak testing using an aerosol photometer and laskin
nozzle aerosol generator. The European Standard for Biosafety cabinets (EN12469:2000) allows the use
of either an aerosol photometer or discrete particle counter for filter testing.
HEPA/ULPA filter leak testing with either an aerosol photometer or discrete particle counter have long
been accepted in cleanroom testing as documented in ISO 14644-3:2005 addressing Test Methods for
Cleanrooms and associated controlled environments and IEST-RP-CC034.1 HEPA and ULPA Filter Leak
Tests. Each method has its advantages and disadvantages, but both are commonly used in cleanroom
and Biosafety cabinet applications globally.
Adding the use of a discrete particle counter to the accepted methods for HEPA/ULPA filter leak testing
for Class II biological safety cabinets in NSF/ANSI 49 would bring the standard into more alignment with
other international standards. In addition, it could make the field certification of Biosafety cabinets less
costly in regions where there are few BSCs or certification. Many of the companies or individuals
servicing cleanrooms and clean air equipment already have a discrete particle counter to classify air
cleanliness and to leak test filters in cleanrooms and other equipment. Requiring the purchase and
maintenance of a costly piece of test equipment simply because NSF has not considered an accepted
methodology hinders public safety by limiting the growth of proper field certification.

Recommendation:
Clearly state what action is needed: e.g., recommended changes to the standard(s) including the current
text of the relevant section(s) indicating deletions by use of strike-out and additions by highlighting or
underlining; e.g., reference of the issue to a Task Group for detailed consideration; etc.
Using the recognized standards, for the consideration of the Joint Committee we have adapted the
current language in Annex F to allow the use of a discrete particle counter to test HEPA/ULPA filters in a
Class II BSC.

F.5 HEPA/ULPA filter leak test (from NSF/ANSI 49-2014)

F.5.1 Purpose
This test determines the integrity of downflow and exhaust HEPA/ULPA
filters, filter housings, and filter mounting frames. The cabinet shall be
operated within 5 ft/min (0.025 m/s) of the nominal set point, with the
exception of the downflow HEPA/ULPA filters on B1 cabinets.
F.5.2 Apparatus

Item No. BSC-2016-7

(For NSF International internal use)
03/2013

Joint Committee Issue Paper

There are two approved methods for HEPA/ULPA filter leak testing; the
aerosol photometer method and the discrete particle counter (DPC) method.
The following devices may be used to carry out HEPA/ULPA filter leak
testing:
F.5.2.1 For the aerosol photometer method, an aerosol photometer with
linear or expanded logarithmic scale shall be used. The instrument
shall be capable of indicating 100% upstream concentration with a
minimum aerosol concentration of 10 g/L of polydisperse
dioctylphthalate (DOP) particles, or an equivalent fluid that provides
the same particle size distribution (e.g., polyalpha olefin [PAO] di[2ethylhexyl], sebecate, polyethylene glycol, and medicinal-grade light
mineral oil)41 produced by the generator described in Annex A,
section A.3.2.2 or equivalent. It shall also be capable of detecting an
aerosol concentration in the downstream equal to 10-5 of the
upstream concentration of the same particles. The sampling rate of
air shall be 1 ft3/min (5 x 10-4 m3/s) 10%. Probe area shall have a
maximum open area of 1.7 in2 (11 cm2) and a minimum dimension of
0.50 in (1.3 cm). The photometer shall be set up in accordance with
the photometer manufacturer's instructions or IEST-RP-CC-013 if
instructions are not provided.
F.5.2.2 For the aerosol photometer method, an aerosol generator of the
Laskin Nozzle type conforming to Annex A, figure A2 or equivalent
shall be used to create an aerosol by flowing air through liquid DOP
or equivalent substitute. When a Laskin nozzle generator is used, the
compressed air supplied to the generator should be adjusted to a
minimum of 20 psi (140 kPa), if using DOP or 23 psi (160 kPa) if
using PAO, measured at the generator manufacturers recommended
location. The nozzles shall be covered with liquid to a depth not to
exceed 1.25 in (31 mm).
F.5.2.3 For the aerosol photometer method, a pressure gauge for the
generator having a maximum range of 0 to 80 psi (0 to 550 kPa) with
resolution and accuracy of 1 psi (7 kPa) calibrated by the
manufacturer or in accordance with the manufacturer's instructions
shall be used.
F.5.2.4 For the DPC method, a discrete particle counter with the capability to
detect particles 0.3 micrometers in diameter and greater and
Item No. BSC-2016-7
(For NSF International internal use)
03/2013

Joint Committee Issue Paper

compliant with ISO 21501-4. The discrete particle counter shall have
an audible alarm that beeps for each particle that is counted and
should have a counting mode that is continuous or at least as long as
the time required to scan a filter. The sampling rate of air shall be 1
ft3/min (5 x 10-4 m3/s) 10%. Probe area shall have a maximum open
area of 1.7 in2 (11 cm2) and a minimum dimension of 0.50 inches (1.3
cm). The particle counter shall be set up in accordance with the
manufacturer's instructions.
F.5.2.5 For the DPC method, a suitable dilution system allowing the
measurement of the upstream concentration. Given the challenge
concentration specification of 12 to 24 million particles per ft3 (420 to
840 million particles per cubic meter), this typically requires a dilution
system of 1:100 or greater.
F.5.2.6 For the DPC method, a suitable challenge aerosol, detectable by the
DPC and in the concentration 12 to 24 million particles per ft3 (420 to
840 million particles per cubic meter) distributed uniformly upstream
of the filter to be tested.
F.5.3 Method of testing HEPA/ULPA filters
F.5.3.1 Filters that can be scanned with an aerosol photometer
a) Turn on the cabinet blower and lights (types A1 and A2 and B2
downflow filter test). Remove the filter diffusers and protective
covers if any are present. Place the generator so the aerosol is
introduced into each cabinet fan upstream of the HEPA/ULPA
filter(s). When the manufacturer has not identified the aerosol
introduction point(s), introduce the aerosol in a manner to ensure
thorough mixing in the cabinet airflow. For example, a Tconnection can be fitted to the aerosol generator output to enable
distribution of challenge into both entrances of a single blower or
entrances of multiple blowers. The manufacturer shall determine
the aerosol introduction point that provides the most uniform
distribution.
b) Turn on the photometer and adjust it in accordance with the
manufacturer's instructions.

Item No. BSC-2016-7

(For NSF International internal use)
03/2013

Joint Committee Issue Paper

c) Determine the aerosol concentration upstream of the
HEPA/ULPA filter.

When the challenged airflow is not contaminated, sample the

aerosol concentration upstream of the HEPA/ULPA filter.

When the challenged airflow is contaminated or when

measuring the upstream concentration is not practical, the
upstream concentration can be calculated. For example, when
DOP is used as the challenge aerosol with a Laskin nozzle
aerosol generator at 20 psi (140 kPa), the following formula
applies:

g/L = 13,500 x number of nozzles / ft3/min of challenged air

NOTE Use of DOP substitutes will require modification of
this formula, unless the photometer is calibrated with the
substitutes to yield results equivalent to those of DOP.
Use of DOP substitutes will also require pressures different
from 20 psig.

Use an aerosol concentration that is at least equal to the

photometric equivalent of 10 g/L of DOP.

d) Set up the photometer to the upstream challenge in accordance

with the photometer manufacturer's instructions to detect leaks
greater than or equal to 0.01% of the upstream concentration.
e) With the nozzle of the probe held not more than 1.0 in (2.5 cm)
from the area being tested, scan the entire downstream side of
the HEPA/ULPA filter(s) and the perimeter of each filter pack by
passing the photometer probe in slightly overlapping strokes at a
traverse rate of not more than 2 in/s (5 cm/s). Separate passes
shall be made around the entire periphery of the filter, along the
bond between the filter pack and frame, and around the seal
between the filter and the device.
f)

Reported values shall be:

upstream aerosol challenge concentration;

Item No. BSC-2016-7
(For NSF International internal use)
03/2013

Joint Committee Issue Paper

method used to report concentration (measured or

calculated);

maximum leak penetration in percent;

method used (scanned or probe tested); and

name of test (HEPA/ULPA filter leak test).

F.5.3.2 Filters that cannot be scanned with an aerosol photometer

a) When a cabinet is ducted so that the exhaust filter cannot be
scanned, it may be leak tested by drilling a hole approximately 0.3
inches (1 cm) in diameter in the duct at a downstream location
that will produce a well-mixed aerosol and inserting the
photometer sampling probe with rigid extension tubing through
the hole.
b) Reported values shall be:

upstream aerosol challenge concentration;

method used to report concentration (measured or
calculated);
maximum leak penetration in percent;
method used (scanned or probe tested); and
name of test (HEPA/ULPA filter leak test).

F.5.3.3 Filters that can be scanned with a discrete particle counter

a) Turn on the cabinet blower and lights (types A1 and A2 and B2
downflow filter test). Remove the filter diffusers and protective
covers if any are present. Place the generator so the aerosol is
introduced into each cabinet fan upstream of the HEPA/ULPA
filter(s). When the manufacturer has not identified the aerosol
introduction point(s), introduce the aerosol in a manner to ensure
thorough mixing in the cabinet airflow. For example, a Tconnection can be fitted to the aerosol generator output to enable
distribution of challenge into both entrances of a single blower or
entrances of multiple blowers. The manufacturer shall determine
the aerosol introduction point that provides the most uniform
distribution.
Item No. BSC-2016-7
(For NSF International internal use)
03/2013

Joint Committee Issue Paper

b) Turn on the particle counter and adjust it in accordance with the
manufacturer's instructions.
c) Determine the particulate concentration upstream of the
HEPA/ULPA filter.

When the challenged airflow is not contaminated, sample the

particle concentration upstream of the HEPA/ULPA filter using
a suitable dilution system to verify a concentration of
detectable particles .

When the challenged airflow is contaminated or when

measuring the upstream concentration is not practical, the
upstream concentration can be calculated from a previous
measurement of volumetric flow and particulate challenge
concentration using the same generator and particle counter.
For example, if C1 (particles per ft3 or m3) was measured in a
flow of Q1 (cfm or m3/s), the concentration of C2 in a flow of Q2
(cfm or m3/s) can be calculated with the following formula:

2 =

1 1
2

Use a detectable particle concentration that is 12 to 24 million

particles per ft3 (420 to 840 million particles per cubic meter).
NOTE: The DPC test parameters identified are derived from
ISO 14644-3 Test Methods, Section B.6.3 Procedure for
installed filter system leakage scan test with a DPC.
472

Note: 472 above from ISO 14644-3 represents the

sample flow rate of 472 cm3/second or 1 cfm
Using the formula for the scan rate above from ISO 14644-3
and IEST-RP-CC034 and inserting the recommended values
specified for challenge concentration, probe width and scan
rate, we can relate Np (the number of particles characterizing
a leak while scanning) directly to the standard leak
penetration and upstream challenge concentration.
Item No. BSC-2016-7
(For NSF International internal use)
03/2013

Joint Committee Issue Paper

Sr = Scan rate shall be 2 inches/sec (5 cm/sec)
Cc = Challenge aerosol concentration upstream of filter in
particles/cm3 shall be 12 to 24 million particles per ft3 (420 to
840 million particles per cubic meter) which yields a minimum
concentration of 420 particles/cm3
PL = standard leak penetration of filter to be tested. The leak
criteria for the photometer test and for a BSC filter leak test
using a particle counter as specified in EN12469 is 0.01% or
0.0001
Dp = probe dimension parallel to the scan direction in cm. Our
specification provides a minimum dimension of 0.5 in (1.3
cm).
Np = expected number of particle counts which characterize the
designated leak while scanning which shall be 5.
472

5 / 420 /3 0.0001 472 3/

1.3
5

d) Set up the particle counter in accordance with the manufacturer's

instructions to detect particles of the required size (0.3
micrometers) and larger.
e) With the nozzle of the probe held not more than 1.0 inch (2.5 cm)
from the area being tested, scan the entire downstream side of
the HEPA/ULPA filter(s) and the perimeter of each filter pack by
passing the photometer probe in slightly overlapping strokes at a
traverse rate of not more than 2 inches/second (5 cm/s). Separate
passes shall be made around the entire periphery of the filter,
along the bond between the filter pack and frame, and around the
seal between the filter and the device. In environments with a
high background of airborne particles care must be taken to
assure accurate results.
A potential leak is identified by an observed particle count Np of 5
particles during the time the probe passes over the defect.
Item No. BSC-2016-7
(For NSF International internal use)
03/2013

Joint Committee Issue Paper

f)

When a potential leak is identified by a particle count during

scanning, the potential leak should be verified by a remeasurement with the probe stationary over the suspect location.
While scanning, any abrupt increase in particle counts should be
cause for investigation. With the minimum concentrations
specified above, a potential leak would be verified with a particle
count of over 290 in a 15 second interval.
NOTE: The DPC test parameters identified are derived from
ISO 14644-3 Test Methods, Section B.6.3 Procedure for
installed filter system leakage scan test with a DPC.
= 472

Npa = Actual number of particle counts which characterize the

designated leak
Tr = Sustained residence time in seconds.
Using the formulas from above and inserting the values
specified, we derive our leak verification value of 300.

290 420

f)

= 472

4723
0.0001
15
3

Reported values shall be:

upstream challenge particle concentration;

method used to report concentration (measured or
calculated);
maximum leak penetration in percent;
method used (scanned or probe tested); and
name of test (HEPA/ULPA filter leak test).

F.5.3.2 Filters that cannot be scanned with a discrete particle counter

a) When a cabinet is ducted so that the exhaust filter cannot be
scanned, it may be leak tested by drilling a hole approximately 0.3
Item No. BSC-2016-7
(For NSF International internal use)
03/2013

Joint Committee Issue Paper

inches (1 cm) in diameter in the duct at a downstream location
that will produce a well-mixed aerosol and inserting the particle
counter sampling probe with rigid extension tubing through the
hole.
b) With the minimum concentrations specified above, a potential
leak of 0.005% would be verified with a particle count of over 145
in a 15 second interval.
NOTE: The DPC test parameters identified are derived from ISO
14644-3 Test Methods, Section B.6.3 Procedure for
installed filter system leakage scan test with a DPC.
Npa = Actual number of particle counts which characterize the
designated leak
Tr = Sustained residence time in seconds.
Using the formulas from above and inserting the values
specified, we derive our leak verification value of 145

145 420

= 472

4723
0.00005
15
3

c) Reported values shall be:

upstream challenge particle concentration;

method used to report concentration (measured or
calculated);
maximum leak penetration in percent;
method used (scanned or probe tested); and
name of test (HEPA/ULPA filter leak test).

F.5.4 Acceptance
F.5.4.1 Filters that can be scanned
Sustained aerosol or particle penetration shall not exceed 0.01% of
the upstream concentration at any point.
F.5.4.2 Filters that cannot be scanned
Item No. BSC-2016-7
(For NSF International internal use)
03/2013

Joint Committee Issue Paper

Sustained aerosol or particle penetration shall not exceed 0.005% of
the upstream concentration.
Supplementary Materials (photographs, diagrams, reports, etc.):
If not provided electronically, the submitter will be responsible to have sufficient copies to distribute to
committee members.
I hereby grant NSF International the non-exclusive, royalty free rights, including non-exclusive,
royalty free rights in copyright; in this item and I understand that I acquire no rights in any
publication of NSF International in which this item in this or another similar or analogous form is
used.

Signature*: David S. Phillips

Company: Thermo Fisher Scientific
Telephone Number:
484-753-3665 E-mail:david.phillips@thermofisher.com
Submission Date:
5/5/2016
Please submit to: Al Rose, arose@nsf.org
*Type written name will suffice as signature

Item No. BSC-2016-7

(For NSF International internal use)
03/2013