1
...:.
!?
ST OF
ORNL P
2387

..
d.
W
.
T
.
".'.
IT
.
fo
PAS
SO
.
1
+
MI
11:25 L4 LE
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS - 1963
SEP 2 2 1966
ORAL-P.2382
conf. 560904.9
LEGAL NOTICE
This report na prepared us wo account of Goverament sponsored work. Neither the Vallad
sulos, bor the Commissi80, bor any person acung on Orball of the Commission:
A. Makes say wvranty or represeautoa, expressed or implied, with respect to the accu-
racy, completeness, or usetilbers of tbe lalormation contained to this report, or what the use
of way lalormation, appesaws, method, or praeus dklosed la Wais report may not latringo
privately owned righus; ur
B. Assumos hay liabilites will respect to the use of, or lor dinugns resulting from the
un of way lalormadas, appariwa, melbod, or process disclosed in this report.
As used to the abovo, "lerson acuag og beball of the Commission' lacludes say on-
ployee or coatractor of the Commission, or employee of suca coatractor, to be extent itu?
such employee or coal actor of Wbe Commission, or employee of such contractor propues,
disseminatos, or provides access to, way information pursuant to his employmeat or coatract
with the Commission, or his employmeat wih such contractor.
REQUIREMENTS FOR FIRE-RESISTANT
HIGH-EFFICIENCY PARTICULATE AIR FILTERS*
C. A. Burchstad
Oak Ridge National Laboratory
Oak Ridge, Tennessee
CFSTI PRICES

RELEASED FOR ANNOUNCEMENT
HCL00; MNISO
IN NUCLEAR SCIENCE ABSTRACTS
Abstract
The establishment of certain basic specification
requirements is essential to the procurement of suitable
high-efficiency particulate air filters with a minimum
of difficulty and cost. Problems have been experienced in
the manufacture and procurement of these filters because
of inconsistent, ambiguous, and conflicting specification
requirements as well as requirements that are not re-
strictive enough. The requirements for high-efficiency
particulate air filters presented in this paper are in-
tended to guide the specification writer toward standards
that will produce uniformity in construction and testing
practices to bring the cost of these units down to reason-
able levels without sacrificing the necessary performance.
Introduction
Specifications written for the procurement of fire-resistant high-efficiency
particulate air (HEPA) filter units used in atoinic energy applications determine
to a great extent the performance capabilities and the cost of the unit procured,
and the requirements for HEPA filters are discussed in this paper to provide a
guide for specification writers. HEPA filters are critical elements in exhaust
and supply air cleaning systems that require an exceptionally high degree of
particulate removal. These filters are very fragile, and many variables in
their construction and handling will affect their operation under the wide range
of environmental conditions to which they may be exposed. Therefore, the
properly written specification will aid in the procurem int of suitable filter
units at minimum cost.
Problems in specification writing fall into two categories: those that
impede manufacture and result in higher costs and procurement delays and those
*Research sponsored by the U. S. Atomic Energy Commission under contract
: with the Union Carbide Corporation.
t-
that can result in inadequate quality of the product. To explore these problems
in specifications and use of HEPA filters, two conferences were held recently by
representatives of users, manufacturers, the Atomic Energy Commissior., the AEC
Quality Assurance Stations, and the Underwriters' Laboratories. The i asults of
these meetings and of practical experience indicate the need for a standard
specification for HEPA filter units to produce uniformity in construction and
testing practices that will bring the cost of these units down to reasonable
levels without sacrificing the necessary performance required for public safety.
Adoption of the requirements and values outlined in the remainder of this paper
may alleviate many of the current problems in the purchase and use of commercially
available HEPA filters.
Performance
The performance of any given type HEPA filter is determined by both non-
destructive and destructive testing. In general, tests of the nondes tructive
type should be run on all filters produced, whereas the destructive type tests
are run only on prototypes of a specific filter unit to determine the suitability
of the design or materials for the wide range of service conditions to witch the
filter may be exposed. These destructive or qualification tests are usually quite
expensive and difficult to conduct, and once made, they need not be repeated in
subsequent orders of filter units of the same materials and construction.
of
Performance Factors Determined by Nondestructive Tests
****
"
-
"
.
Nondestructive tests are made on a filter unit to determine its efficiency
and pressure drop, penetration, and whether or not it leaks around the frame.
Efficiency and Pressure Drop. Particle removal efficiency and the pressure
drop across the filter are two of the most important properties of any HEPA
filter unit; the first is a measure of its primary function and the second has
a direct bearing on its operating cost. By definition, a HEPA filter is one
that (1) has a minimum removal efficiency of 99.97% when tested with a thermally
geparated 0.3-micron-diameter particle dioctyle phthalate smoke and (2) has a
max:' num clean filter pressure drop of 1.0 in, H,O when operated at rated capacity.
Nondestructive tests to confirm these properties are made by the manufacturer
and, when specified, by the AEC Quality Assurance Station, following Edgewood
Arsenal procedures.
The figure of 99.97% is a reasonable compromise between necessary manu-
facturing tolerance and protection for the user; most of today's filters test
at substantially higher values. For a given filter design, pressure drop is a
direct function of flow capacity or of face velocity. The figure of 1.0 in. HO
is arbitrary, but it is specified to permit direct comparison between filter
units of different manufacture or design or of tests on the same filter made at
different times or places. It 13 use does not preclude operation of the filter
at either higher or lower flows or face velocities.
Penetration. Two-flow testing is recommended by the AEC to disclose some
pinholes and leaks that cannot be found by full-flow testing or visual inspection.a
Pinholing is a media de fact that permits unfiltered air to pass through the filter
even in units that show full-flow test efficiencies as high as 99.99%.3 Since
penetration of pinholed filters is independent of particle size, 4 particles with
diameters of several microns can escape through the filter. A statistical
analysis of filter test results showed a 95% probability of the existence of
significant pinholing if the difference in penetration between the full-flow test
and a 20%-flow test exceeded 0.01%;6 this is borne out by the work of Ad ley and
Knuth.
Inclusion of a requirement for two-flow testing, at full flow and at 20%
flow, with a maximum permissible difference in the absolute value of penetration
of 0.01% is recommended in specifications for filter units with a capacity of
500 cfm and larger. Two-flow testing of smaller units is not practical since
the low flow velocities at 20%-flow testing make the results questionable.
Frame Leakage. A requirement for frame leak testing is recommend ed because
of the significant number or "frame leakers" discovered during Quality Assurance
Station testing. Frame leaks can result from improper manufacturing prncesses or
from properties of the materials themselves. These are discussed in the section
on materials and construction.
Performance Factors Determined by Destructive Tests
Destructive or qualification tests are made on prototype filter units to
determine their resistance to fire and hot air, moisture resistance, pressure
loading, and dust holding capacity. Qualification of a given design or set of
materials by one manufacturer does not confer qualification of a similar design
er manufacturer because differences in methods of con-
struction could result in differences in performance.
Resistance to Fire and Hot Air Qualification tests for fire and hot air
are made by Underwriters' Laboratories in accordance with Standard UL-586.8
Qualification includes a hot sic test at a temperature of 700°F for five minutes
while the filter is being operated at its rated capacity and a spot flame test
in which the flame from a Bunsen burner is directed against the upstream face
of the filter for five minutes while the filter is being operated at its rated
capacity. Although the standard permits a reduction of 3% in filtration effici-
ency following hot air testing, the results of tests usually show a reduction
of less than 0.3%.9 The spot flame test is a test of combustibility rather than
a test of the effect on filtration performance or filtration efficiency. The
test will breach the filter and is successful if no sustained burning takes place
on the downstream face of the filter.
When specified, Underwriters' Laboratories qualification is evidenced by
the UL label on the filter unit. It is not necessary in writing specifications
to go to great length in spelling out fire resistance, flame spread ratings of
materials, cracking or shrinkage of adhesives, or similar details. A11 of
these can be covered by the single requirement for UL labeling.
Underwriters' Laboratories qualification is not a certification of filter
efficiency nor is it a substitute for Quality Assurance Station or other per-
formance testing. It only certifies that the materials and construction meet
the requirements of UL-586, and that materials and construction of units furnished
on the order are identical to those of the unit originally tested. A valuable
service of the Underwriters' Laboratories program is spot checking of manu-
facturers to ensure compliance of labeled filter production, and part of this
check is to make sure that there has been no change in materials or construction.
Moisture Resistance. Moisture resistance has two aspects : resistance to
plugging of the filter when exposed to low-quality stean or fog containing
water droplets and resistance to saturation that would reduce the strength of
the media or separators below some tolerable minimum. Tests for moisture resisto
ance are still under development and will be discussed at weet ings being held
concurrently with this conference. The results of those meetings will be
reported at a later date.
Pressure Loading. To withstand occasional overpressure and operation under
high dirt loadings with a reasonable factor of safety, HEPA filters must be able
to withstand a pressure drop of 10 in. H, O across the filter while operating at
full rated capacity under both dry and molot conditions without rupture or loss
in filtration efficiency. As with moist ure testing, no suitable tesc has been
developed to date. Uniform specification of both moisture and pressure loading
resistance is recommended as soon as suitable tests become available. This will
make possible a standard qualification procedure that can perhaps be included in
a later version of UL-586.
Dust Holding Capacity. Dust holding capacity is a measure of potential
filter life, and it is a direct function of media area. It is not desirable to
specify this property indirectly by stipulating a minimum media area, a minimum
number of separators, or a maximum flow velocity through the media itself because
these may conflict with other requirements. For example, the nunber of separators
will change in accordance with the type and thickness of media and separators. A
minimum separator count based on a unit with 0.7015-in.-thick aluminum separators
might be impossible to meet if a unit with 0.016-in.-thick as bestos separators
were specified in a purchase order. Conversely, a separator count based on
asbestos separators would permit too loose a filter pack should a unit with
aluminum separators be ordered.
Specification of dust holding capacity as a performance requirement is
recommended. The recommended minimum loadings to produce a 2-in. H; O pressure
drop while operating at rated capacity, using a National Bureau of Standards
Cottrell precipitator test dust are tabulated below.
-
-
..
.
-
- -
.
Minimum
Dust Loading
(grams)
Filter Size
(scfm)
.
.
-
-
.--.-
25
50
.-
125
500
1000
75
160
400
1650
3000
Both test dusts and minimum recommended dust loadings are still under investigation,
The distribution of particle sizes in the National Bureau of Standards test dust
may vary considerabl; from batch to batch, and a more uniform dust, such as that
proposed by the Air Filter Instituie, may be proposed when more is known about its
loading characteristics in HEPA filters. For the present, the above values are
adequate,
.
.
..
.. ..
..
.
.
?.
*
.
Materials and construction
*
**
-
isien
e
r
e
The commercially available open-face HEPA filters covered by this guide
conform to a rather uniform design and construction. They usually consist of a
filter pack, made by pleating fiber glass paper back and forth over corrugated
aluminum or asbestos frame, sealed in a plywood or cadmium-plated steel frame
with a suitable aihesive. This configuration provides a unit with maximum
filtration area (about 230 fts in a 1000-cfm unit) and minimum frontal area and
volume (4 ft2 and 4 ft3, respectively, in a 1000-cfm unit). Materials and con-
struction have a direct bearing on performance, resistance to damage and deteri-
orat ion under operating conditions, and cost. Other materials may be used to
meet special requirements but should be specified with care.
S
Frames
The strength and rigidity of the frame directly affect the ability of the
filter to withstand mounting bolt loa'zings and racking and warping that could
damage the filter pack or its seal. The usual frame materials are 3/4-in.
exterior-grade A-A plywood or No. 16 U.S. gage cadmium-plated carbon steel.
Plywood should be specified to be in accordance with either Commercial
Standard CS-45 or CS-122,1°, 11 which define the requirements for exterior and
type grading. Lesser grades than A-A should not be specified because of the
tendency of manufacturers' personnel to turn the poor side in for appearances
sake, which could prevent an adequate seal between the frame and the filter pack.
Wood frames should have rabbeted corners and be axis embled by double nailing on.
doubig screwing (i.e., fasteners through both surfaces of the joint) using coated
box naiis, galvanized flat-head screws, or galvanized screw nails. Mat-formed
wood-particle board has been proposed as an alternate for plywood, It offers
economies in filter construction, but preliminary tests have shown sufficient
porosity to cause frame leaks. This material is still under investigación, and
the results of this investigation will be reported at a later date.
Cold-rolled carbon steel sheet is preferred to hot-rolled sheet because it
is flatter and is held to closer dimensional and surface tolerances. Steel frames
are usually made up of four panels assembled with cadmium plated rivets, bolts,
or drive screws. Frame panels should be cadmium plated after completion of all
shearing, preforming, and welding operations to ensure proper coaring of sheared
edges and weld areas. Minimum coating of Type Os in accordance with ASTM A16513
is recommended. Without this restriction, the mailufacturer could legitimately
furnish a plate so thin as to be almost useless. À Type OS plate is 7.6 microns
thick. Frames should have double-turned flanges on both faces; single-turned
flanges are too weak to withstand the bolt loads often applied to seal these
filters into their mountings. Abutting sections of the flange face should have
no more than a 1.32-in. gap to assure proper seating of the gasket. Gaps as
large as 1/8 in. have beer. observed. On both wood and steel frames, the mating
surfaces of corner joints should be thoroughly coated with adhesives before
assembly to ensure a tight joint.
If stainless steel frames are required, the gauge and assembly requirements
for carbon steel should be required for the stainless steel frames except that
the fasteners should also be stainless steel. Since stainless steel is speci-
fied for its corrosion resistance, it is advisable to specify a grade and con-
formance to ASTM A240.13
Frame Dimensions. Five sizes of HEPA filters have been widely accepted as
standard and these are tabulated below.
Capacity
(s cfm)
Face
Depth
Dimensions : Less Gaskets
(in.)
(in.)
25
50
125
500
1000
8 by 8
8 by 8
12 by 12
24 by 24
24 by 24
3 1/16
5 7/8
5 7/8
5 7/8
11 1/2
The use of these standard sizes is recommended for new construction because higher
costs will result from using special sizes due to manufacturing inefficiencies,
stocking problems, and procurement delays. It is not necessary to design a
- system to use filter units exactly matched to system flow. If the next larger
standard size 18 specified, pressure drops will be lower (with resultant fan and
duct saving), longer filter life will be obtained (with resultant savings in both
procurement and maintenance costs), and excess capacity will be provided for the
Special sizes may be required because of inattention to details in the design
and construction of filter banks and housings. One user is burdened with the
necessity of buying 23 1/2- by 23 1/2-in. filter units for one system because
insufficient spicing was provided between mounting bolts. Although a standard-
size unit will go into the available space, the least amount of moisture will
cause swelling that jams the unit in the frame. There has been a recent trend
toward the use of very large filter units in certain applications with face
dimensions up to 4 by 6 ft. Uniess heavily reinforced, these units are structur-
aily weak and are subject to pinholing and cracking of the media and the seal
between the filter pack and frame caused by racking and warping during shipping
and handling.
Dimensional Tolerances. Specifications for face riimension tolerances range
from + 1/32 in. in 2 ft, which is beyond the accuracy of a production woodworking
shop, to + 1/8 in., which is so loose that a user coulc' have trouble in seating
the filter in a tight mounting. The recommended frame tolerances are tabulated
below.
Size of Filter Unit
25, 50, 125 scfm
500, 1000 scfm
Face Dimensions
Depth
Diagonals
+0, - 1/16 in.
+ 1/16 in., -0
Equal within 1/16 in.
+0, - 1/8 in.
+ 1/16 in., -0
Equal within 1/8 in.
The plus zero tolerance on the face dimension provides for interchangeability in
housings or tight mountings without selective fitting. The minus zero tolerance
on depth ensures a minix.im gasket deflection in mountings witn toggle or cam-
operated ciamping devic.. that have a fixed amount of deflection.
Because the condition of gasket surfaces is critical to proper sealing of
-
-
-
in. total indicator reading (TIR) as measured with one face resting on a truly
flat surface, such as a machinist's surface plate. Gasket surfaces should be
square with the sides of the frame to within + 29.
The usual separator materials are asbestos (containing no more than 5%
combustible material and having a basic weight of 4 lbs/100 ft) and aluminum
(0.0015-in.-thick minimum, H-19 temper, preferably a solution hardening alloy
since work hardening alloys anneal and soften at fairly low temperatures).
However, asbestus has practically no moisture resistance, and aluminum has
neither corrosion nor high-temperature resistance. Aluminum will fail nearly
xplosively at temperatures above 1000°F. Polyvinyl chloride and stainless
steel have both been used for corrosion resistance. However, polyvinyl chloride
cannot be fire rated and is too soft even at room temperature, and stainless
steel is prohibitively expensive. Specially-treated asbestos papers are under
development by all four HEPA filter manufacturers and several paper manufacturers.
These show promise of excellent resistance to moisture after exposure to con-
centrated reagents, fire, and steam. Two materials have met all requirements of
a tentative specification test, 14 and should be available upon establishment of
a formal qualification agency and completion of I fire tests, ...
Media
Investigations to improve media tensile and wet strength, to increase media
bulk (to provide better protection against pinholing dun to separator vibration
during shipping and operation), and to develop a hydrogen-fluoride-resistant
media are in progress. Results of these investigations will be reported at a
future date.
Recognizirg a number of drawbacks in the current version of liilitary Standard
MIL-F-51079,46 the following requiremenc is suggested as an interim specification.
Glass or gla88-asbestos paper at least 0.015 in. thick containing no
more than 5% combustible material in accordance with MIL-F-51079
except cross machine tensile strength shall be 2.5 lbs/in. paragraph
3.2.10 (pH) shall be deleted; and paragraph 3.2.6.2 shall be changed
to say that the minimum tensile strength after folding shall be no
less than 50% of the original tensile strength.
Spot patching to repai: holes or tears in media cause "nort spots" when the
filter is exposed to hot air or fire and should not be permitted. The defect
should be cut out and the media repaired by splicing. Where two pieces of media
are to be joined by splicing, each piece should be coated with adhesive for 1
to 2 in. across the full width of t' i sheet. A limitation of one or two splices
in a 500 or 1000 cfm filter unit is unreasonable because the manufacturer has no
control over where he finds defects or the end of a roll. Such limitations
increase costs due to the excessive waste involved, and a limit of four or five
splices may be reasonable if such a requirement is considered necessary. A
restriction on the number of splices in small.er units may be desirable because
of flow restriction caused by splicing. Because of smaller quantities of media
involved, waste would be correspondingly less and could be held to within reason.
Adhesives
The adhesive used in a HSPA filter must retain sufficient strength and
resiliency to maintain its sealing integrity under exposure to flowing hot, cold,
dry, or moist air, steam and condensed moisture, some small amount of working of
the frame, shear forces set up by pressure loading, and other conditions to which
the filter may be exposed in service. Adhesives can therefore only be qualified
by destructive tests of the entire filter. In addition to tests by Underwriters'
Laboratories, tests are made by the Atomic Energy Commission. Currently available
adhesives that meet the requirements of both UL-586 and AEC testing are tabulated
below.
Minnesota Mining & Manufacturing Company
Pittsburgh Plate Glass Company
Goodyear Pliobond
Talon Adhesives Company
No. EC-2155
No. 2-743
No. HT-30-FR
Ceman-TAC NB-38A
Cemen-TAC NB-38B
Cemen-TAC NB-380
In addition to these commercially available adhesives, a proprietary foamed-in-
place polyurethane has proved very successful. Because the choice of adhesive
is 80 critical to performance of the filter, specification of one of these
materials or of an identified alternate specifically approved by the buyer is
recommended.
Workmansliip
Loose packing and kinked or nonperpendicular pleats and separators can
result in inedia damage due to separator vibration and working of the media
during shipping and operation. Although not so common today, they still give
enough trouble that a requirement limiting them may be.desirable. Some degree
of kinking and nonperpendicularity of separators must be allowed for manufac-
turing tolerance and has not been found to be harmful by DOP testing. Reason-
able limits are a maximum kink of 1/4 in. from a line drawn from end to end of
the separator and a maximum deviation of + 1/4 in. from a perpendicular drawn
from the opposlte end of the separator. Greater deviation is evidence of poor
workmanship.
Loose packing is not so easily defined. A reasonable test is to place
one's hand :irmly against the face of the filter and attempt to move it laterally;
any motion of the separators is cause for rejection. This test is not suitable
with polyvinyl chloride, however, because polyvinyl chloride is 80 soft that the
corrugations will flatten and permit separator movement. Separators should extend
at least 1/8 in. from the pleats to give adequate protection to the media.
Gaskets
Gasket selection is the subject of a study by F. E. Adley that is to be
reported at this conference. Numberous materials have been used for gaskets,
including fiber glass mat, cork-rubber and several grades of neoprene. Until
the results of Mr. Adley's research can be evaluated, closed-cell expanded
neoprene grade SCE-43 in accordance with ASTM B N 5616 is recommended. Fiber
glass mat is too fragile and will leak from edge to edge regardless of how
much it is compressed. Cork-rubber, though specified to be 20-40 durometer,
is usually 60 or harder, and it requires very high clamping-bolt loads to seal
the filter. Neoprene softer than SCE-43 is too soft. ASTM D1056 measures
hardness in terms of compression because the hardness of sponge materials
cannot be accuratcly determined with a durometer. Grade SCE-43 regires a
load of 11 1/2 lb/in.2, + 2 lb, to produce 25% compression.
Face Guards
Although recommended by the AEC for several years, face guards are not
required by most users. HEPA filters can be very easily damaged during handling
and installation or by materials scraped across them while mounted in a bank or
housing. I have recently had the privilage of visiting air cleaning facilities
at a large number of atomic energy installations throughout the county, and in
nearly every one where face guards were not used, some degree of face damage to
installed filters was evident. The savings gained by eliminating face guards
can quickly be lost in the cost of a few damaged filter units.
The least expensive face guard is 4-mesh galvanized hardware cloth conforme
ing to Comercial Standard CS 132.17 Edges should be firmly embedded in adhesive
(the Quality Assurance Stations reported numerous loose face guards) and project-
ing wires at the edges must not penetrate the media or extend outwards where
they could be a hazard to personnel.
Face guards are only recommended for 1000-cfm filter unics or larger, and
they should be specified to be on both faces. Smaller filte: units are light
enough and manageable enough that the probability of damage by persons handling
them is relatively low.
.
. . . . -
IVET
-
-
.
.
.--
--
--
S
--
-
-
*.
-
-
'
ALAM
Acceptance, Packaging, and Shipping
Inspection and nondestructive acceptance testing of HEPA filter units for
use in the U. S. atomic energy program are carried out by the AEC Quality Assurance
Stations at Oak Ridge, Tennessee for sitec east of the Mississippi River and
Hanford in Richland, Washington. Acceptance of fire-resistant units is based on
presence of the Underwriters' laboratories label, inspection for shipping damage
or obvious defects, and successful completion of performance tests. When speci-
fied in the specification or purchase order, testing includes two-flow and frame
leak testing
Acceptance
Altnough in-place testing of filter systems by the user is highly recommended,
it is not a suitable substitute for acceptance testing of filter units by the
Quality Assurance Stations. The hot smoke (i.e., thermally generated diociyl
phthalate used in Quality Assurance Station testing is a uniform dispersion of
0.3-micron-diameter particles that permits determination of the basic removal
efficiency for small particles. QAS testing also provides a valuable check on
rmance tests by the manufacturers and ensures uniformity of testing procedures.
Neither of these can be done with the cold smoke (air-generated dioctyl phthalate)
used for in-place testing. Because cold smoke is a polydisperse mixture of
particle sizes ,* penetration values produced by cold smoke testing are distorted
by a preponderance of large particles. Therefore, very high efficiency readings
from cold smoke testing tell nothing about the removal efficiency for the very
small particles against which the filter was designed to protect. Small particle
removal efficiency can only be determined by hot smoke testing. Because of the
high cost of hot smoke testing facilities, this service is provided through the
stations at Hanford and Oak Ridge.
Title to filter units normally passes to the buyer upon successful completion
of inspection and testing. This does not preclude subsequent inspection by the
buyer for shipping damage or conformance to materials and construciion requirements,
or special performance requirements not covered by Quality Assurance Station testing.
Packaging
Zach filter unit should be legibly and permanently marked with the manufacturer's
name or symbol, the filter serial number, the actual percent penetration at rated
capacity (as determined by manufacturer's test), the clean fil:er pressure drop
(as determined by manufacturer's test), the air flow capacity, and an arrow showing
the direction of air flow during test. Marking should be on the top of the filter
unit with separators vertical and should be used as a guide by the buyer's personnel
when installing the filter. Filters should always be installed with the separators
vertical.
*Eight percent of the particles produced by a NRL Type II aerosal generator
were reported to have diameters larger than 1.88 microns, and 77% had diameters
larger than 0.59 microns. 18
10
The serial number should be printed with figures at least 1/2 in. high with
a sequencing numbering machine to avoid duplicate numbering or misnumbering.
Other characters may be either machine or hand printed, but they should be at
least 3/8 in. high for legibility. Labels, particularly typewritten, should not
be permitted for marking.
Cartons used for shipping HEPA filters, in addition to meeting Interstate
Commerce Commission regulations, must provide adequate shock protection and must
users go to some length in specifying packaging requirements. An excellent carton
is that described in U.S. Patent No. 3,057,468,13 which is reported by the Quality
Assurance Stations to be one of the nost satisfactory both from the standpoint of
protection and from the standpoint of safe handling and reuse. This carton has
been made available to all manufacturers for government and government-contractor
orders without royalty.
Tops of cartons should be sealed with tape and not glue or stapes because
the cartons will be reused by the QAS and in many instances by the buyer for
packaging dirty filters removed from his system. The exterior of the carton should
be marked with the manufacturers name or symbol, the filter description, the
buyers purchase order number, and FRAGILE and THIS SIDE UP legends. Filters
should always be packed in cartons with pleats and faces vertical, and cartons
should be handled, stored, and shipped with the faces vertical. A statement in
the specification outlining these requirements is recommended.
bility of the carrier and not the manufacturer unless inadequate packaging has
been provided because the Quality Assurance Stations repack filters in the same
cartons. If such damage is frequent, the buyer should look to his packaging and
shipping methods rather than to elimination of Quality Assurance Sčation testing.
Shipment of filters by rail is not recommended because of the humping and
jarring normally encountered. If shipped by truck, it is advisable tu require
that the filters be shipped in the same truck or trailer from the QAS to the
buyer's receiving dock, with no transshipment. Personnel handling filters'in
the buyer's receiving department should be instructed in proper handling of
HEPA filters, see USAEC report TID-7023,30 and they should be required to unload
all shipments. Cartons should be shipped and stored in a vertical position and
stacked no more than three high.
References
1.
Filter Testing Penetraneter Instruction Manuals 076, 0107, and Q117,
Edgewood Arsenal, U. S. Army. These manuals replace MIL-STD-282 and will
eventually be replaced by an American Standard.
H. Gilbert, "Panel B - Round Table Discussion," Proceedings of the Eighth
Air Cleaning Conference Held at Oak Ridge National Laboratory October 22-25,
1963, USAEC Report TID-7677, p. 439.
F. E. Adley and D. E. Anderson, "The Effect of Holes on the Performance
Characteristics of High-Efficiency Filters," Proceedings of the Eighth Air
Cleaning Conference Held at Oak Ridge National Laboratory October 22-25,
1963, pp. 494-507.
4.
J. W. Thomas, "Aerosol Penetration Through Pinholed Filters," Health Physics
2: 667-673, Pergamon Press, 1965.
5. C. A. Burchsted, Oak Ridge National Laboratory, unpublished data, July 1966.
6. F. E. Adley, Hanford Occupational Health Foundation, personal communication,
1966.
7. R. H. Knuth, "Performance of Defective High Efficiency Filters," Am. Ind,
Hys. Assoc. J., 26: 593-600, November-December 1965.
8. Standard UL-586, High Efficiency Air Filter Units, Underwriters' Laboratories,
Chicago, Illinois, June 1964.
L. Horn, "Panel B - Round Taile Discussion, "Proceedings of the Eighth Air
Cleaning Conference Held at Oak Ridge National Laboratory October 22-25,
1963, USAEC Report TID-7677, p. 443.
Commercial Standard CS-45-64, Standard for Douglas Fir Plywood, U. S.
Department of Commerce, 1964.
11. Commercial Standard cs-122-64, Standard for Western Softwood Plywood, u. s.
Department of Commerce, 1964.
12. ASTM A165-55, "Specification for Electrodeposited Coatings of Cadmium on
Steel," ASTM Standards 1966, Vol. 1, American Society for Testing and
Materials, 1966.
ASTM A240-63, "Specification for Chromium and Chromium-Nickel Stainless
Steel Plate, Sheet, and Strip for Fusion Welded Unfired Pressure Vessels,"
ASTM Standards 1966, Vol. 3, American Society for Testing and Materials,
1966.
14. C. A. Durchsted, "Qualification Test for Moisture and Corrosion Resistant
Separator Materials," Tentative ORNL Standard AR-2.01, Oak Ridge National
Laboratory, June 1966.
15. Military Specification MIL-F-51079, Filter Medium, Fire Resistant, High
Efficiency, U. S. Army, April 6, 1.962.
ASTM D1056-66, "Specification for Sponge and Cellular Rubber Products, "ASTM
Standards 1966, Vol. 28, American Society for Testing and Materials, 1966.
17. Commercial Standard CS-132-46, Standard for Hardware Cloth, U. s. Department
of Commerce, 1946.
U. S. Naval Research Laboratory Report No. 5929, July 26, 1963.
19. A. R. Allan, Jr., Package for Filters, U. S. Patent No. 3,057,468, October 9,
1962.
H. Gilbert and J. H. Palmer, High Efficiency Particulate Air Filter Units;
Inspection, Storage. Handling Installation, USAEC Report TID-7023,
August 1961.
.
.
.
-
END
-
.
DATE FILMED
10/24 / 66







.
..