877 KB

877 KB
NOTICE 1
Document ID:
Document Date:
FED-STD-209E
11-29-2001
Per the DOD Document Automation and Production Service:
Overview Title:
AIRBORNE PARTICULATE CLEANLINESS CLASSES IN CLEANROOMS
AND CLEAN ZONES (S/S BY ISO14644-1 AND ISO14644-2)
Status:
Cancelled
[METRIC]
I?ED-STD-209E
Serkember 11, 1992
SUPERSEDING
FED-STD-209D
June 15, 1988
FEDERAL STANDARD
AIRBORNE PARTICULATE CLEANLINESS CLASSES
IN CLEANROOMSAND CLEAN ZONES
This Standard is approved by the Commissioner, Federal
Supply Service, General Services Administration, for the
use of all Federal agencies.
e
DISTRIBUTION STATEMENT A: Approved for
public release; distribution is unlimited..
FSC 3694
I
FED-sTD-209E
September 11, 1992
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1.1
Scope . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1.2
Limitations
. . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
. . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
.
.
.
.
.
.
.
.
.
.
1
1.
●
e
0
SCOPE AND LIMITATIONS
2.
REFERENCED DOCUMENTS
3.
DEFINITIONS
3.1
Airborne particulate cleanliness class
3.2
Aniaokinetic
sampling . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
1
3.3
Calibration
. . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3.4
Clean zone
. . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3.5
Cleanroom.
. . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3.5.1
As-built cleanroom
(facility) . .
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3.5.2
At-rest cleanroom
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3.5.3
Operational cleanroom
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3.6
Condensation nucleus counter (CNC)
.
-. .
.
.
.
.
.
.
.
.
.
2
3.7
Discrete-particle
3.8
Entrance plane
3.9
Isoaxial
(facility)
(facility)
counter (DPC ) .
.
.
.
.
.
.
.
.
.
.
.
.
.
2
. . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
. . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3.10
Isokinetic sampling . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3.11
Monitoring
. . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3.12
Nonunidirectional
airflow . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
3.13
Particle
. . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
3.14
Particle concentration.
. . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
3.15
. .
Particle size . . . . . . ,
.
.
.
.
.!
.
.
.
.
.
.!
.
.
.
3
.
i
FED-sTD-209E
September 11, 1992
3.16
Student’s t statistic . . . . . . . . . .
.
.
.
.
.
.
.
.
.
3
3.17
Udescriptor
. . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
3
3.18
Ultrafine particles
. . . . . . . . . . .
.
.
.
.
.
.
.
.
.
3
.
. . . . . . . . .
.
.
.
.
.
.
.
.
.
3
. . . . . .
.
.
.
.
.
.
.
.
.
3
. . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
3
AIRBORNE PARTICULATE CLEANLINESS CLASSES AND U DESCRIPTORS
.
.
.
4
.
.
.
4
3.19
Unidirectional
3.20
Upper confidence limit (UCL)
3.21
Verification
4.
airflow
4.1
Classes listed in Table I . . . . . . . .
4.1.1
Measurement
at particle sizes listed in Table I . . . .
.
.
4
4.1.2
Measurement
at alternative particle sizes . .
4.2
Provision for defining alternative airborne
particulate
.
.
.
.
.
.
.
.
.
.
.
.
4
.
.
.
.
.
.
.
4
. .
.
.
5
. . . . . . . . . . . .
.
.
6
. . .
.
.
6
cleanliness classes. . .
.
.
6
. . . . . . . . . . . . . . .
.
.
7
cleanliness classes . . . . . . .
Airborne particulate cleanliness classes
Table I.
4.3
.
.
Provision for describing ultrafine particle
concentrations
Nomenclature
(U descriptors)
for airborne particle concentrations
Format for airborne particulate
Format for U descriptors
VERIFICATION
AND MONITORING OF AIRBORNE PARTICULATE CLEANLINESS
. . . . .
7
.’. . . . . . . . . . . . . . . . . . . . . . .
7
Verification
Frequency.
7
of airborne particulate
Environmental
test conditions
cleanliness
. . . . . . . . . . . . . . .
7
. . .
7
factors . . . . . . . . . . . . . . . . . . .
8
Status of cleanroom or clean zone during verification
Environmental
.
Particle counting.
. . . . . . . . . . . . . . . . . . . .
ii
.
I
FED-sTD-209E
September 11, 1992
o
5.1.3.1
Sample locations and number: unidirectional
airflow . . . .
5.1.3.2
Sample locations and number: nonunidirectional
5.1.3.3
airflow
8
. .
9
Restrictions on sample locations
. . . . . . . . . . . . .
9
5.1.3.4
Sample volumeandsampling
. . . . . . . . . . ...9
5.1.3.4.1
Single sampling plan for classes in Table I
5.1.3.4.2
Single sampling plan for alternative classes or
-.
I,.
particle sizes
a
e
. . . . . . .
9
. . . . . . . . . . . . . . . . . . . . . .10
5.1.3.4.3
Single sampling plan for U descriptors
5.1.3.4.4
Sequential sampling plan
. . . . . . . . . . . . . . ...11
5.1.4
Interpretation ofthedata
. . . . . . . . . . . . . ...11
5.2
Monitoring of airborne particulate cleanliness
5.2.1
Xonitoringplan.
. . . . . . . . . . 11
5.2.2
Particle counting for monitoring
5.3
Methods and equipment for measuring airborne
. . . . . . 11
. . . . . . . . . . . . . . . . . . . . .11
particle concentrations
9
time
. . . . . . . . . . . . . 12
. . . . . . . . . . . . . . . . . . 12
5.3.1
Counting particles’5 micrometers and larger . . . . . . . . 12
5.3.2
Countingparticlessmallerthan 5 micrometers. . . . . . . 13
5.3.3
Countingultrafineparticles . . . . . . . . . . . . . . . 13
5.3.4
Limitationsof particlecountingmethods . . . . . . . . . 13
5.3.5
Calibrationof particlecountinginstrumentation. . . . . 14
5.4
Statisticalanalysi8 . . . . . . . . . . . . . . . . . . . 14
5.4.1
Acceptancecriteria for verification
. . . . . . . . . . . 14
5.4.2
Calculations to determine acceptance
. . . . . . . . . . . 14
5.4.2.1
Average particle concentration at a location
5.4.2.2
Meanoftheaverages
. . . . . . . 14
. . . . . . . . . . . . . . . . . . .14
iii
FED-STD-209E
September 11, 1992
●
5.4.2.3
Standard deviation of the averages . . . . .
.
.
.
.
.
.
.
15
5.4.2.4
Standard error of the mean of the averages
.
.
.
.
.
.
.
.
15
5.4.2.S
Upper confidence limit (UCL) . .
.
.
.
.
.
.
.
.
15
.
.
15
.
.
.
.
UCL factor for 95% upper confidence limit
Table II.
5.4.2.6
.
Sample calculation
.
.
.
.
.
.
.
.
.
.
15
.
.
.
.
.
.
.
.
.
.
.
15
.
.
.
.
.
.
.
.
.
.
.
.
16
.
.
.
.
.
.
.
.
.
.
.
.
16
. . . . . . .
.
,.
. . . . .
.
.
.
.
. . . . . .
.
6.
RECOMMENDATION
FOR CHANGES
7.
CONFLICT WITH RJ3FERENCED DOCUMENTS
8.
FEDERAL AGENCY INTERESTS
APPENDIX A
AND SIZING AIRBORNE PARTICLES
. . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 17
. . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
. . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 18
. . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
A60 .
Calibration of the microscope
.
.
.
.
.
.
.
.
.
.
.
.
.
. 20
A70.
Counting and sizing particles by optical microscopy
.
.
.
. 22
A80 .
Reporting
A90 .
Factors affecting precision and accuracy
A1O.
Scope.
A20.
Summary of the method
A30.
Equipment
A40.
Preparation of equipment
A50.
Sampling the air
. . . . . . . . . . .
.
.
.
.
.
19
.
.
.
.
.
.
.
. 23
. .
.
.
.
.
.
.
. 23
.
●
APPENDIX B
OPERATION OF A DISCRETE-PARTICLE
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 24
. . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 24
. . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 25
Apparatus and related documentation
.
.
.
.
.
.
.
.
.
.
.
.
B1O .
Scope and Limitations
B20.
References
B30.
Summary of method
B40 .
iv
26
.
FED-sTD-209E
September 11, 1992
0
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
28
.
. . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
30
.
. . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
30
sampling
B50.
Preparations
B60 .
Sampling . . .
B70 .
Reporting
. .
c
,-
SAMPLING
ISOKINETIC AND ANISOKINETIC
Clo.
Scope. . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
31
C20.
Reference
. . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
31
C30.
Background
. . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
31
C40.
Methods
. . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
31
. . . . . . . . .
.
.
.
33
Probe inlet diameters
Figure Cl.
for isokinetic sampling, v =Vo.
Probe inlet diameters
Figure C.2.
for isokinetic sampling, v =Vo.
(English units)
. . . . . . .
.
.
.
.
.
33
Contours of sampling bias, C/CO =
Figure C.3.
.C50.
(metric units)
0.95, 1.05 . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
34
. . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
35
Example
APPENDIX D
METHOD FOR MEASURING THB CONCENTRATION OF ULTRAFINE PARTICLES
. . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
36
References
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
36
Apparatus
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
36
efficiency of a DPC used to verify the U descriptor
. . .
37
Determining the concentration of ultrafine particles
. . .
37
D1O.
Scope
D20.’
D30 .
.
Figure D.1.
D40.
Envelope of acceptability for the counting
v
FED-STD-209E
September 11, 1992
APPENDIX E
RATIONALE FOR THE STATISTICAL RULES USED IN FED-STD-209E
E1O.
Scope.
. . . . . . . . . . . . . . . . . . . . . . . . . . 38
J
. 38
E20.
The statistical rules . . . . . . . . . . . . . . ..’..
E30.
Sequential sampling . . . . . . . . . . . . . . . . . . . . 40
E40 .
Sample calculation to determine statistical validity of a
verification
. . . . . . . . . . . . . . . . . . . . . . .40
APPENDIX” F
SEQUENTIAL SAMPLING: AN OPTIONIiL METHOD FOR VERIFYING THE COMPLIANCE OF AIR
TO THE LIMITS OF AIRBORNE PARTICULATE CLEANLINESS CLASSES M 2.5 AND CLEANER
F1O.
Scope . . . . . . . . . . . . . . . . . . . . . . . . . . .42
F20.
References
. . . . . . . . . . . . . . . . . . . . . . . .42
F30.
Background
. . . . . . . . . .. . . . . . . . . . . . . .
F40.
Method
. 42
. . . . . . . . . . . . . . . .. . . . . . . . . . . 42
Figure F.1.
Observed count, C, vs. expected count, E,
for sequential sampling . . . . . . . . . . . . . . . . .
Table F.1.
Upper and lower limits for time at which
C counts should arrive
F50.
Examples
F60 .
Reporting.
. 43
. . . . . . . . . . . . -----. . .
. 44
. . . . . . . . . . . . . . . . . . . .“. . . . . 45
. . . . . . . . . . . . . . . . . . . . . . . . 45
.-
APPENDIX G
SOURCES OF SUPPLEMENTAL
. . . . . . . . . . . . . . . . . . . . . . . .
G1O .
Scope..
G20.
Sources of supplemental information
vi
:
INFORMATION
. 46
. . . . . . . . . . . . 46
FED-STD-209E
September 11, 1992
1.
ScoDe and limitations.
1.1 -.
This document established standard classes, and provides for
alternative classes, of air cleanliness for cleanrooms and clean zones based on
specified concentrations of airborne particles.
It prescribes methods for
verifying air cleanliness and requires that a plan be established for monitoring
air cleanliness.
It also provides a method for determining and describing
concentrations (U descriptors) of ultrafine particles.
1.2 Limitations.
The requirements of this document do not apply to equipment
or supplies for use within cleanrooms or clean zones. Except for size
classification and population, this document is not intended to characterize the
physical, chemical, radiological, or viable nature of airborne particles.
No
universal relationship has been established between the concentration of airborne
particles and the concentration of viable airborne particles.
In addition to the
need for a clean air supply that is monitored for total particulate contamination
and that meets established limits, special requirements are necessary for
monitoring and controlling other forms of contamination.
2.
Referenced documents.
2.1 Box, George E. P., Hunter, William G., and Hunter, J. Stuart, Statistics for
Experimenters, John Wiley & Sons, New York, 1978.
‘a
2.2 Hinds, W. C., Aeroeol Technolouv : Properties. Behavior, and Measurement of
Airborne Particles, John Wiley & Sonsr New York (1982).
2.3 FED-STD-376,
Government.
Preferred Metric Units for General Use bv the Federal
The International System of units (S1) is preferred.
In the event of a conflict
between S1 and U. S. customary units, S1 units shall take precedence.
3.
Definitions.
3.1 Airborne particulate cleanliness class. The level of cleanliness specified
by the maximum allowable number of particles per cubic meter of air (per cubic
foot of air), shown for the class in Table I, as determined by the statistical
methods of 5.4. The name of the class in S1 units is taken from the logarithm
(base 10) of the maxtium allowable number of particles, 0.5 #m and larger, per
cubic meter.
The name of the class in English (U.S. customary) units is taken
from the maxim~
allowable number of particles, 0.S gm and larger, per cubic
foot .
3.2 Anisokinetic samlinq.
The condition of sampling in which the mean velocity
of the flowing air stream differs from the mean velocity of the air entering the
inlet of the sampling probe. Because of particle inertia, anisokinetic sampling
can cause the concentration of particles in the sample to differ from the
concentration of particles in the air being sampled.
FED-sTD-209E
September 11, 1992
3.3 Calibration.
Comparison of a measurement standard or instrument of unknown
accuracy with another standard or instrument of known accuracy to detect,
correlate, report, or eliminate by adjustment any variation in the accuracy of
the unknown standard or instrument.
3.4 Clean zone. A defined space in which the concentration of airborne
particles is controlled to meet a specified airborne particulate cleanliness
class.
3.5 Cleanroom.
A room in which the concentration of airborne particles is
controlled and which contains one or more clean zones.
3.5.1 As-built cleanroom (facilitv~. A cleanroom (facility) that is complete
and ready for operation, with all services connected and functional,. but without
equipment or operating personnel in the facility.
A cleanroom (facility) that is complete,
3.5.2 ~.
with all services functioning and with equipment installed and operable or
operating, as specified, 1 but without operating personnel in the facility.
3.5.3 Operational cleanroom {facilitv). A cleanroom (facility) in normal
operation, with all services functioning and with equipment and personnel, if
applicable, present and performing their normal work functions in the facility.
3.6 Condensation nucleus counter (CNC)_. An instrument for counting small
airborne particles, approximately 0.01 pm and larger, by optically detecting
droplets formed by condensation of a vapor upon the particles.
3’.7 Discrete- particle counter (DPC)_. An instrument, such as an optical particle
counter or a condensation nucleus counter, capabl”e of resolving responses from
individual particles.
3.8 Entrance Diane. A plane perpendicular to the unidirectional airflow located
immediately upstream of the region of interest (typically the work area unless
otherwise specified) and having the same dimensions as the cross section of the
clean zone perpendicular to tfie direction of the airflow.
3.9 Isoaxial. A condition of sampling in which the direction of the airflow
into the sampling probe inlet is the same as that of the unidirectional airflow
being sampled.
3.10 Isokinetic sam~linq.
The condition of isoaxial sampling in which the mean
velocity of the air entering the probe inlet is the same as the mean velocity of
the unidirectional airflow at that location.
3.11 Monitoring.
The routine determination of airborne particle concentrations,
as well as other relevant conditions, in cleanrooms and clean zones.
lWhen terms such ‘as “shall be specified,” “as specified,” etc. are used without
further reference, the degree”of control needed to meet requirements will be
specified by the user or contracting agency.
2
●
PED-sTD-209E
September 11, 1992
m
3.12 Nonunidirectional airflow. Airflow which does not meet the definition of
unidirectional airflow; previously referred to as “turbulent” or “non-laminar”
airflow.
3.13 Particle.
An object of solid or liquid composition, or both, and generally
between 0.001 and 1000 pm in size.
3.14 Particle concentration.
of air.
The number of individual particles per unit volume
3.15 Particle size. The apparent maximum linear dimension of a particle in the
plane of observation as seen with a microscope, or the equivalent diameter of a
particle detected by automatic instrumentation.
The equivalent diameter is the
diameter of a reference sphere having known properties and producing the same
response in the sensing instrument as the particle being measured.
3.16
Student’s t statistic.
The distribution:
t = [(sample mean) - (population mean)]/[standard
error of the sample mean]
obtained from sampling a normal (Gaussian) distribution. Tables of critical
values are available in statistics texts (see 2.1).
@
3.17 U descriptor. The maximum allowable concentration (particles per cubic
meter of air) of ultrafine particles.
The U descriptor serves as an upper
confidence limit.or as the upper limit for the location averages, or both, as
appropriate
e. U descriptors are independent of airborne particulate cleanliness
classes, and may be specified alone or in conjunction with one or more airborne
particulate cleanliness classes.
3.18 Ultrafine ~articles.
Particles in the size range from approximately
0.02 pm to the upper limit of detectability of the DPC described in Appendix D.
Ultrafine particles are operationally defined by the relationship for counting
efficiency vs. particle size of Appendix D.
3.19 Unidirectional airflow. Airflow having generally parallel streamlines,
operating in a single direction, and with uniform velocity over its cross
section; previously referred to as “laminar” airflow.
.
3.20 Urmer confidence limit (UCLL. lin upper limit of the estimated mean which
has been calculated so that, in a specified percentage of cases, its value
exceeds the true population mean, both means having been sampled from a normal
(Gaussian) distribution.
In this Standard, a 95% UCL is used.
3.21 Verification.
The procedure for determining the compliance of air in a
cleanroom or clean zone to an airborne particulate cleanliness class limit or a
U descriptor, or both, as specified.a
Iwhenterns such as “shall be specified,” “as specified”
@
etc. are used without
further reference, the degree of control needed to meet requirements will be
specified by the user or contracting agency.
3
FED-STD-209E
September 11, 1992
Verification of
4. Airborne particulate cleanliness classes and U descrirkors.
air cleanliness, in accordance with section 5 of this Standardr utilizes a system
of classification based upon specified limits.
This section defines standard classes of air cleanliness, each having specific
concentrations of airborne particles in specific particle size ranges (see
Table I). Provisions are also made for defining standard classes based upon
alternative particle sizes, and for defining alternative (nonstandard) classes.
In addition, a basis is provided for describing air cleanliness in terms of
concentrations (U descriptors) of ultrafine particles.
A system of nomenclature
is given.
suitable for describing all classes and U descriptors
4.1 Classes listed in Table I. For the airborne particulate cleanliness
classes listed in Table I, verification of air cleanliness shall be performed by
measurement at one or more of the particle sizes listed in Table I or at other
specified particle sizes, as follows:
4.1.1 Measurement at particle sizes listed in Table I. Verification shall be
performed by measurement at one or more of the particle sizes listed for the
class in Table I, as specified,l and shall be reported using the format described
in 4.4.1. The airborne particulate cleanliness class is considered met if the
particle concentration measurements for the specified size or sizes are within
the limits given in Table I, as determined by the statistical analysis of 5.4.
4.1.2 Measurement at alternative particle sizes. Verification may be performed
by measurement at particle sizes other than those listed in Table I, with the
following limitation: The alternative particle size or sizes selected must be
within the range of sizes listed for the indicated class in Table I. The
airborne particulate cleanliness class is considered met if the particle
concentration measurements for each selected alternative size do not exceed the
limit given in Table I for the next larger particle size,. as determined by the
statistical analysis of 5.4. Verification shall be reported using the format
described in 4.4.1.
4.2 Provision for defininu alternative airborne Qarticulate cleanliness classes.
Classes other than those shown in Table I (for example, Classes M 2..2, M 4.3, and
M 6.4 (Classes 5, 600, and 70 000)) may be defined when special conditions
dictate their use. The name for an alternative class shall be based on the
concentration limit specified for particles 0.5 pm and larger, in the same manner
as the classes listed in Table I. Concentration limits for other particle sizes
shall be in the same proportions as those of the next cleaner class in Table I;
these limits can be calculated by using the appropriate equation in the footnote
under Table I. Similarly; for classes cleaner than Class M 1 or Class 1, the
concentration limits at particle sizes other than 0.5 pm shall be in the same
proportions as those of Class M 1 or Class 1.
~When terms such as “shall be specified,” “as specified,” etc. are used without
further reference, the degree of control needed to meet requirements will be
specified by the user or contracting agency.
4
●
. ..
.
I
FED-sTD-209E
September 11, 1992
TAELE I
AIREORNE PARTICULATE CLEANLINESS CLASSES
Class limits are given for each clase name. The limits designate specific
concentrations (particles per unit volume) of airborne particles with sizes
equal to and larger than the particle sizes shown”
Cta9a
lids
0.1pm
Clasa
Namc-
SI
--
M 1.5
1
M2
M 2.5
10
M3
volume
Un.ila
volume
unitS
volume
Units
volume
Units
(m~
(m~
(m
(m?
(&)
[d
(d
2.14
30.9
0.875
10.0
0.283
7.50
likl
3.00
35.3
1.00
8.75
(!P)
9.91
7$.7
265
3500
99.1
757
21.4
309
75.0
1060
30.0
353
10.0
87.5
1ooo
28.3
&
--
350
2650
35000
991
7570
214
30S0
26500
750
10600
300
3s30
100
75700
2140
30900
875
1000O
283
35300
1000
247
100000 2830
618
17.5
353000 low
2470
70.0
Iwoooo 28300
6180
low
10000
Mb
M6.5
--
2.83
100
(b
12a
MS
M 5.5
(m
3s.0
M4
M 4.5
S#rn
1240
100
M 3S
0.5Jml
Vohnnoullh
350
Ml
0.3#m
0.2fun
100m
M7
7.00
175
3530000 looocm
24700. 700
10000000283000
61800 1750
~he class limits shown in Table I are defined for classification purposes only
and do not necessarily represent the size distribution to be found in any
particular situation.
‘Concentration limits for intermediate classes can be calculated, approximately,
from the following equations:
--
particles/n?
= 10”(0.5/d)n
where M is the numerical designation of the class based on S1 units,
and d is the particle size in micrometers, or
particles/fts = Nc(0.5/d)u
where Nc is the numerical designation of the class based on English
(u. s. customary) units, and d is the particle size in micrometers.
m
nor
naming and describing the classes, S1 names and units are preferred;
however, English (U.S. customary) units may be used.
5
FED-STD-209E
September 11, 1992
When expressed in S1 units, the numerical designation of the class is derived
from the logarithm (base 10Z with the mantissa truncated to a single decimal
place) of the maximum allowable number of particles, 0.5 pm and larger, per cubic
meter of air. When expressed in English (U. S. customary) units, the numerical
designation of the class is derived from the maximum allowable number of
particles, 0.5 pm and larger, per cubic foot of air.
(a)
For alternative classes less clean than Class M 4.5 (Class 1000),
verification shall be performed by measurement either in the particle
size range 0.5 pm and larger or in the particle size range 5 pm and
larger, or both, as specified.1
(b)
For alternative classes cleaner than Class M 4.5 (Class 1000) but
less clean than Class M 3.5 (Class 100), verification shall be
performed by measurement in one or more of the particle size ranges:
0.2 pm and larger, 0.3 pm and larger, and 0.5 pm and larger, as
specified.1
(c)
For alternative classes cleaner than Class M 3.5 (Class 100),
verification shall be performed by measurement in one or more
of the particle size ranges: 0.1 pm and larger, 0.2 pm and larger,
0.3 pm and larger, and 0.5 pm and larger, as specified.1
4.3 Provision for describing ultrafine particle concentrations (U descriptors).
A U descriptor, if specified,l shall be used to express the concentration of
ultrafine particles as defined in 3.17. The U descriptor may supplement the
class definition or may be used alone. The format for U descriptors is described
in 4.4.2.
4.4
Nomenclature
●
●
for airborne ~article concentrations.
4.4.1 Format for airborne Darticulate cleanliness classes.
expressed by using the format “Class X (at Y pm),” where:
Classes shall be
X represents the numerical designation of the airborne particulate
cleanliness class; and
Y represents the particle size or sizes for which the corresponding
particle concentration (class) limits are specified.
For example:
“Class M 2.5 (at 0.3 pm and 0.5 Mm)” describes air with not more
than 1060 particles/m3 of a size 0.3 pm and larger, nor more
than 353 particles/m3 of a size 0.5 pm and larger.
“Class 100 (at 0.5 pm)” describes air with not more than 100
particles/ft3 of a size 0.5 pm and larger.
Iwhen terns such
as “shall
be specified,” “as specified,” etc. are used.without
further reference, the degree of control needed to meet requirements will be
specified by the user or contracting agency.
—
6
a
FED- STD-209E
September 11, 1992
.
4.4.2 Format for.U descriptors.
A U descriptor may be used alone or as a
supplement to the specification of an airborne particulate cleanliness class.
Specifying a particle size for U descriptors is unnecessary, since the lower
cutoff for ultrafine particles is determined by the equipment used (see 3.18 and
Appendix D).
U descriptors
shall be expressed by using the format “U(x),” where:
x is the maximum allowable concentration
of ultrafine particles.
(particles per cubic meter of air)
For example:
“U(20)” describes air with not more than 20 ultrafine particles/xt?.
“Class M 1.5 (at 0.3 P),
U(2000)” describes air with not more
than 106 particles/r?of a size 0.3 pm and larger, and not more
than 2000 ultrafine particles/~.
5.
Verification
and monitoring of airborne na*iculate
cleanliness.
5.1 Verification of airborne Darticulate cleanliness.
Verification, the
procedure for determining the compliance of air in a cleanroom or clean zone to
an airborne particulate cleanliness class limit or a U descriptor, or both, as
defined in section 4, shall be performed by measuring the concentrations of
airborne particles under the conditions set forth in 5.1.1 through 5.1.4. The
particle size or sizes at which the measurements are to be made for verification
shall be specified, using the appropriate format as described in 4.4.
5.1.1 Freouency.
After initial verification, tests shall be performed at
periodic intervals, or as otherwise specified.l
Verification of air cleanliness shall be
5.1.2 Environmental test conditions.
accomplished by measuring particle concentrations under specified’ operating
conditions, including the following.
The status of
5.1.2.1 Status of cleanroom or clean zone durinu verification.
the cleanroom or clean zone during verification shall be reported as “as-built,”
“at-rest,” “operational ,“ or as otherwise specified.i
%4hen terms such as “shall be specified,” “as specified,” etc. are used without
further reference, the dearee of control needed to meet requirements will be
specified by the user or ~ontracting agency.
7
FED-STD-209E
September 11, 1992
5.1.2.2 Environmental factors. Measurements and observations of applicable
environmental factors related to the cleanroom or clean zone during verification
Such factors may include, but are not limited to, air
shall be recorded.
velocity, air volume change rate, room pressurization, makeup air volume,
unidirectional airflow parallelism, air turbulence> air temperature, humidity or
dew point, and room vibration.
The presence of equipment and personnel activity
should also be noted.
5.1.3 Particle countinq. Verification of air cleanliness in cleanrooms and
clean zones shall be performed in accordance with the appropriate particle
counting method or methods in 5.3, as specified.t Appropriate sampling locations
and sampling plan”shall be selected from the following subparagraphs.
For unidirectional
5.1.3.1 Sample locations and number: unidirectional airflow.
airflow, the sample locations ‘shall be uniformly spaced throughout the clean zone
at the entrance plane, unless otherwise specified,l except as limited by
equipment in the clean zone.
The minimum number of sample locations required for ,verification in a clean zone
with unidirectional airflow shall be the lesser of {a) or (b):
(a) S1 units: A/2.32
where A is the area of the entrance plane in mz
English
(U. S. customary) units: A/25
where A is the area of the entrance plane in ft2
(b)
S1 units: A x 64/(10”)Os
where A is the area of the entrance plane in mz, and M is the
S1 numerical designation of the class listed in Table I
English
(U. S. customary) units: A/(NC)Os
where A is the area of the entrance plane in ft2, and Nc is the
numerical designation of the class, in English (U. S. customary)
units, listed in Table I
The number of locations shall always be rounded to the next higher integer.
‘when terms such as “shall be specified,” “as specified~” etc. are used without
further reference, the degree of control needed to meet requirements will be
specified by the user or contracting agency.
8
●
i
FED-STD-209E
September 11, 1992
e
5.1.3.2 Samu le locations and number: nonuni.directional airflow. For
nonunidirectional airflow, the sample locations shall be uniformly spaced
horizontally, and as specified vertically, throughout the clean zone, except as
limited by equipment within the clean zone.
The ❑inimum number of sample locations required for verification
with nonunidirectional airflow shall be equal to:
.
in a clean zone
S1 units: A X 64/(10”)U
where A is
the floor area of the clean zone in s?, and M is
the S1
numerical designation of the class listed in Table I
English
(U. S. customary) units: A/(Nc)Os
where A is the floor area of the clean zone
in ft=, and Nc is the
numerical designation of the class in English (U. S. customary) units
listed in Table I
The number of locations shall always be rounded to the next higher integer.
e
5.1.3.3 Restrictions on sanmle locations. No fewer than two locations shall be
sampled for any clean zone. The sample locations shall be uniformly spaced
throughout the clean zone except as ltiited by equipment within the clean zone.
At least one sample shall be taken at each of the sample locations selected
(see 5.1.3.1 or S.1.3.2). More than one sample may be taken at each location,
and different numbers of samples may be taken at different locations, but a total
of at least five samples shall be taken in each zone. Sampling at more locations
than the required minimum will result in greater precision in the mean of the
location averages and, when applicable, its upper confidence limit.
sample volume and sam~lina time. The volume of air sampled and time of
sampling shall be determined in accordance with the applicable paragraph below.
5.1.3.4
5.1.3.4.1
Sinale sanmlina Dlan for classes in Table I. Each sample of air
tested at each location shall be of sufficient volume such that at least 20
particles would be detected, if the particle concentration were at the class
limit, for each specified particle size. The following formula provides a means
of calculating the minimum volume of air to be sampled as a function of the
m
‘When
terms such as “shall be specified,” “as specified,” etc. are used without
further reference, the degree of control needed to meet requirements will be
specified by the user or contracting agency.
9
FED-STD-209E
September 11, 1992
●
number of particles per unit volume listed in the appropriate cell of Table I:
Volume = 20 parti.cles/[class limit (particies/volume)
from Table I]
The volume of air sampled shall be no less than 0.00283 m3 (0.1 ft3), and the
results of the calculation of the sample volume shall not be rounded down.z
A larger sample volume will decrease the variation between samples, but the
Sample
volume should not be so large as to render the sampling time impractical.
volumes need not be identical at all locations; however, the particle
concentration shall be reported in terms of particles per cubic meter (per cubic
foot) of air regardless of the sample volume. The volume of air sampled shall
also be reported.
Sampling larger volumes than the required minimum will result
in greater precision in the mean of the location averages and its upper
confidence limit.
.
.
The sampling time is calculated by dividing the sample volume by the sample flow
rate.
5.1.3.4.2 Sincrle sanmlina plan for alternative classes or particle sizes. The
minimum sample volume required for verifying the compliance of air to other class
limits, as defined in 4.2, shall be the volume determined for the next cleaner
class listed in Table I, in accordance with the procedure described in 5.1.3.4.1.
The minimum sample volume for verification by measurement at alternative particle
sizes, as described in 4.1.2, shall be the volume determined for the next larger
particle size shown in Table I, in accordance with the procedure described in
5.1.3.4.1.
The sampling time is calculated by dividing the sample volume by the sample flow
rate.
Other considerations concerning sample volume, as detailed in 5.1.3.4.1, also
apply in these situations.
‘Example:
The minimum sample volume for Class M 2.5 (at 0.5 pm) [Class 10 (at 0.5 pm)]:
volume = 20 particles/(353
= 0.0567 m3
particles/m3
or
volume = 20 particles/(10
particles/ft3
= 2.00 ft3
10
●
I
FED-STD-209E
Sept&mber 11, 1992
a
5.1.3.4.3 Sinale samplinq plan for u descrirkors.
The sample volume required
for verifying the concentration of ultrafine particles shall be the volume of air
sufficient to permit at least 20 particles to be detected at the specified
U descriptor.
The minimum volume, in cubic meters, shall be calculated by
dividing 20 by the U descriptor.
The results of this calculation shall not be
rounded down, and in no case shall the volume be less than 0.00283 ~.
The sampling time is calculated by dividing the sample volume by the sample flow
rate.
5.1.3.4.4
Sequential samplina Plan. As an alternative method for verifying the
compliance of air to the limits of airborne particulate cleanliness Classes M 2.5
and cleaner (Classes 10 and cleaner), the sequential sampling plan described in
Appendix F may be used (Sequential Sampling: An Optional Method for Verifying the
Compliance of Air to the Limits of Airborne Particulate Cleanliness Classes M 2.5
and Cleaner).
The advantage of sequential sampling is the potential to reduce
significantly the sample volume at each location and, consequently, to reduce
sampling times.
5.1.4 Inte rpretation of the data. Statistical evaluation of particle
concentration measurement data shall be performed, in accordance with 5.4, to
verify compliance of air to airborne particulate cleanliness class limits or
U descriptors, or both. If a sequential sampling plan is used, the data analysis
described in Appendix F shall be used.
o
5.2 Monitoring of airborne Particulate cleanliness.
After verification,
airborne particulate cleanliness shall be monitored while the cleanroom or clean
zone is operational, or as otherwise specified.1~ Other environmental factors,
such as thos”e listed in 5.1.2.2, may also be monitored as specified to indicate
trends in variables that may be related to airborne particulate cleanliness.
5.2.1 Monitoring nlan. A monitoring plan shall be established based on the
airborne particulate cleanliness and the degree to which contamination must be
controlled for protection of process and product, as specified.1
.
m
tem8 such as “shall be specified,” “as specified,” etc. are used without
further reference, the degree of control needed to meet requirements will be
specified by the user or contracting agency.
Iwhen
3For monitoring purposes only, determining the extent to which particles are
contaminating surfaces may be accomplished by allowing airborne particles to
deposit on test surfaces and then counting them by appropriate methods.
The
relationship between airborne. and deposited particles, however, is complex.
Although the concentration of airborne particles in ambient air is an important
variable influencing the deposition of these particles, it is not the only
variable; unfortunately, the magnitudes of many of the other variables may either
be unknown ox not easily measured.
Therefore, while the rate of particle
deposition on surfaces can be a suitable monitor for airborne particulate
cleanliness, an unambiguous relationship cannot be given.
11
FED-STD-209E
September 11, 1992
The plan shall specify the frequency of monitoring, the operating conditions, and
the method of counting particles.
The number of locations and the number and
volume of samples, as well as the method used for interpreting the data, shall
also be specified.
●
5.2.2 Particle countina for monitoring.
Particle counting for the monitoring
plan shall be performed using one of the methods in 5.3, as specified.1 Particle
concentration measurements shall be made at selected locations throughout the
clean zone, or where cleanliness levels are especially critical, or where higher
particle concentrations have been found during verification.
-.
The
5.3 Methods and eau inment for measurina airborne Particle concentrations.
method and equipment to be used for measuring airborne particle concentrations
The
shall be selected on the basis of the particle ”size or sizes specified.
methods in the following paragraphs are suitable for verifying the compliance of
air to airborne particulate cleanliness class limits or U descriptors, as
Other particle
appropriate, and may also be used for monitoring air cleanliness.
counting methods or equipment, or combinations of other methods and equipment,
may be used if demonstrated to have accuracy and repeatability equal to or better
than these methods and equipment.
Equipment used to determine the concentration of airborne particles shall be
properly maintained in accordance with the manufacturer’s instructions and
periodically calibrated, as specified.1
5.3.1 Countinq particles 5 micrometers and laraer. The concentrations of
particles in the range 5 pm and larger shall be determined by using the
procedures in Appendix A (Counting and Sizing Airborne Particles Using Optical
Microscopy).
●
Alternatively, a discrete-particle counter (DPC) may be used if the procedures
described below for sample acquisition, handling, and measurement are satisfied.
The counting efficiency of the DPC for particles larger than 5 pm shall be stated
in accordance with the procedures of Appendix B (Operation of a Discrete-particle
Counter). The DPC shall be operated to count only those particles 5 pm and
larger.
Whichever method is used, the probe inlet dimensions, sample flow rates, and
probe orientations should be selected to permit isokinetic sample acquisition.
Isokinetic sampling is preferred, but if it cannot be achieved, an estimate of
sampling bias shall be obtained by using the procedures of Appendix C (Isokinetic
and Anisokinetic Sampling).
Even when the probe inlet ”faces directly into the
airflow (isoaxial sampling), artificial enrichment or depletion of the ambient
particle concentration can occur at the inlet if the velocity of the airflow into
the inlet differs from the velocity of the airflow in the immediate vicinity of
the probe.
Formulas are available for calculating the effects of such
lWhen terms such as “shall be specified,” ‘*asspecified,” etc. are used without
further reference, the degree of control needed to meet requirements will be
specified by the user or contracting agency.
12
@
FED-STD-209E
September 11, 1992
anisokinetic sampling upon observed measurements of particle concentrations
(see, for example, 2.2 and Appendix C). For particles 0.5 pm and smaller, these
artificial enrichments and depletions in concentration can be shown to be less
than 5% and can be ignored. For particles 5 pm and larger, however, when the
predicted artificial change in concentration exceeds 5%, the projected increase
or decrease should be reported and the correction applied to the data before
comparison with airborne particulate cleanliness class limits.
The concentrations of
5.3.2 Countinu narticles smaller than 5 micrometers.
particles smaller than 5 pm shall be determined by using a DPC in accordance with
the procedures “of Appendix B. Particle size data shall be reported in terms of
equivalent diameter as calibrated against reference standard particles.
m
As mentioned (see 5.3.1), the bias resulting from anisokinetic sampling is less
than S% for particles 0.5 ym and smaller, but can be substantial for particles
5 pm and larger. However, if verification is to be performed at one or more
particle sizes in the range between 0.5 pm and 5 @
(that is, intermediate sizes
not listed in Table I), the likelihood of bias from anisokinetic sampling
increases with increasing particle size, and a correction for anisokinetic
sampling may be necessary.
5.3.3 Countina ultrafine Darticles.
The concentrations of ultrafine particles
shall be determined by using the procedures of Appendix D (Method for Measuring
the Concentration of Ultrafine Particles).
Discrete-particle counters with
5.3.4 Limitations of Particle countincl methods.
unlike designs or operating principles may yield different data when used to
sample air at the same location. Even recently calibrated instruments of like
Caution should be used when comparing
design may show significant differences.
measurements from different instruments.
DPC’S shall not be used to measure particle concentrations or particle sizes
exceeding the upper limits specified by their manufacturers.
.
Since the sizing and counting of particles by optical microscopy defines size on
the basis of a “longest dimension,” while DPC’S define size on the basis of
“equivalent diameter,” particle concentration data obtained from the two methods
may not be equivalent and therefore shall not be combined.
13
FED-STD-209E
September 11, 1992
All instruments shall
5.3.5 Calibration’of Particle countina instrumentation.
be calibrated against known reference standards at regular intervals using
accepted procedures, as specified.1 Calibration may include, but is not limited
to, airflow rate and particle size. Calibration with respect to particle size
shall be carried out for each size measured in verification.
Collection and statistical analysis of airborne
5.4 Statistical analvsis.
particle concentration data to verify the compliance of air to specified airborne
particulate cleanliness class limits or U descriptors shall be performed in
This statistical analysis deals
accordance with the following subparagraphs.
only with random errors (lack of precision), not errors of a nonrandom nature
(bias) such as erroneous calibration.
If a sequential sampling plan is used, the data shall be treated in accordance
with the analysis described in Appendix F.
A rationale for the statistical methods used in this Standard is given in
Appendix E.
5.4.1 Acceptance criteria for verification.
The air in a cleanroom or clean
zone shall have met the acceptance criteria for an airborne particulate
cleanliness class (see Table I for standard limits) or U descriptor when the
averages of the particle concentrations measured at each of the locations fall at
Additionally, if the total number of
or below the class limit or U descriptor.
locations sampled is less than ten, the ‘mean of these averages must fall at or
below the class limit or U descriptor. with a 95% UCL.
5.4.2
Calculations
●
to determine acceptance.
5.4.2.1 Avera~e particle concentration at a location. The average particle
concentration, A, at a location is the sum of the individual sample particle
concentrations, G, divided by the number of samples taken at the location, N, as
shown in equation 5-1. If only one sample is taken, it is the average particle
concentration.
A=
(Cl+%+
(Equation 5-1)
... +C~)/N
5.4.2.2 Mean of the averaues. The mean of the averages, M, is the sum of the
individual averages, ~, divided by the number of locations, L, as shown in
equation 5-2. All locations are weighted equally, regardless of the number of
samples taken.
M=
(AI+AZ+
(Equation 5-2)
... +A~)/L
lWhen terms such as “shall be specified,” “as specified,” etc.’ are used without
further reference, the degree of control needed to meet requirements will be
specified by the user or contracting agency.
14
●
i
FED-sTD-209E
September 11, 1992
5.4.2.3 Standard deviation of the averaues. The standard deviation of the
averages, SD, is the squqre root of the sum of the squares of differences between
each of the individual averages and the mean of the averages, (~ - 24)2,divided
by the number of locations, L, minus one, as shown in equation 5-3.
.
(~-M)2+(~-M)a+. ..+(A~-M)2
L-1
SD=
.
(Equation 5-3)
5.4.2.4 Standard error of the mean of the averaqes. The standard error, SE, of
the mean of the averages, M, is determined by dividing the standard deviation,
SD, by the square root of the number of locations, as shown in equation 5-4.
~E=sQ
(Equation 5-4)
a
5.4.2.5 Wmer confidence limit fUCL1. The 95% UCL of the mean of averages, M,
is determined by adding to the mean the product of the appropriate UCL factor
(see Table II) and the standard error, SE, as shown in equation 5-5.
(Equation 5-5)
UCL = M + (UCL Factor x SE)
TABLE 11.
UCL FACTOR FOR 95% UPPER CONFIDENCE LIMIT
No. of locations, L
95% UCL factor
“
2
3
4
5
6
7
8
9
6.31
2.92
2.35
2.13
2.02
1.94
1.90
1.86
>9”
NA .
-hen the number of locations is greater than 9, the calculation of a
UCL is not required (see 5.4.1).
.
5.4.2.6
sample calculation.
A sample calculation is given in Appendix E.
6. Recommendation for chanues. When a Federal agency considers that this
Standard does not provide for its essential needs, written request for changing
this Standard, supported by adequate justification, shall be sent to the General
Services Administration (GSA). This justification shall explain wherein the
Standard does not provide for essential needs. The request shall be sent to the.
General Services Administration, General Products Commodity Center, Federal
Supply Service, Engineering Division (7FXE), 819 Taylor Street, Fort Worth, TX
76102. The GSA will determine the appropriate action to be .take.nand will nOtifY
the requesting agency.
FED-STD-209E
September 11, 1992
7. Conflict with referenced documents. Where the requirements stated in this
Standard conflict with any document referenced herein, the requirements of this
Standard shall take precedence.
The nature of such conflicts shall be submitted
in duplicate to the General Services Administration, General Products Commodity
Center, Federal Supply Service, Engineering Division’ (7FXE), 819 Taylor Street,
Fort Worth, TX 76102.
8.
Federal aaencv interests.
Department of Commerce
Department of Defense, Office of the Assistant Secretary of Defense
(Installations and Logistics)
Army
Navy
Air Force - Custodian - 99
- Reviewer
- 84
Department of Energy
Department of Health and Human Services
Department of Transportation
General Services Administration
National Aeronautics and Space Administration
Nuclear Regulatory Commission
,-.
.
16
FED-STD-209E
September 11, 1992
APPENDIX A
COUNTING AND SIZING AIRBORNE PARTICLES USING OPTICAL MICROSCOPY
A1O . -.
This appendix describes methods for determining the concentration
of particles S pm and larger in cleanrooms and clean zones. By collecting the
particles on a membrane filter and counting them using optical microscopy, their
concentration in the air which is sampled can be determined.
A20. Summaxw of the method.
1-
A20.1 Description.
Using vacuum, a sample of air is drawn through a membrane
filter. The rate of flow is controlled by a limiting orifice or by a flowmeter;
thus, the total volume of air sampled is determined by the sampling the.
The
kmbrane
filter is subsequently examined microscopically to determine the number
of particles 5 @ and larger collected from the sample of air.
A20.2 Alternatives to optical microscopy.
Image analysis or projection
microscopy may replace direct optical microscopy for the. sizing and countkg of
particles, provided the accuracy and reproducibility equal or exceed that of
direct optical microscopy.
e
A20.3 Acceptable samDlinu Procedures.
Two acceptable procedures for sampling
air for particles are described in this Appendix: (a) the Aerosol Monitor Method,
=d (b)-the Open Filter Holder Method.
A30.
EauiDment.
A30.I
Eauinm ent common to both methods.
A30.1.1 Microscope.
Binocular microscope with ocular-objective combinations
capable of 100- to 250-fold magnifications.
A combination should be chosen so
that the smallest division of the ocular reticle, at the highest magnification,
is less than or equal to 5 P.
The objective used at the highest magnification
should have a numerical aperture of at least 0~25.
A30.I.2 Ocular reticle. A 5- or 10-mm scale with 100 divisions or a micrometer
eyepiece with a movable scale.
.
A30.1.3
Staae micrometer.
A conventional stage micrometer with scale
gradations of 0.01 to 0.1 mm per division.
A30.1.4
External illuminator.
A30.1.5 Vacuum source. Capable of maintaining a vacuum of 67 k.pa (9.7 lb/inch2)
while pumping at a rate of at least 0.00047 #/s (1 fts/min).
A30.1.6
Timer. 60-minute ranae.
A30.1.7
train.
Flow’meter or limitina orifice.
Calibrated in line with the vacuum
@
17
FED-STD-209E
September 11, 1992
A30.1.8
Manual counter.
Petri slides with covers for the storage of
A30.1.9
Filter storaqe holders.
membrane filters after use and during counting.
A3O.1.1O
Rinse fluid. Distilled or deionized water subsequently
through a membrane having pores 0.45 pm or smaller.
A30.1.11
A30.2
Forcens.
filtered
Flatr with underrated tips.
Equ iDment specific to Aerosol Monitor Method.
A30.2.1 Aerosol monitors.
Dark, pore size 0.8 pm or smaller, with imprinted
grid, and white (for contrast when counting dark particles), pore size 0.8 pm or
smaller, with imprinted grid.
A30.2.2
A30.3
Aerosol adapter and tubinq.
EquiDment specific to Open Filter Holder Method.
A30.3.1
Filter holder.
A30.3.2
Membrane filters. Dark, pore size 0.8 pm or smaller, with imprinted
grid, and white (for contrast when counting dark particles), pore size 0.8 pm or
smaller, with imprinted grid.
Optional eau ipment.
A30.4
A30.4.1
Imaqe analvzer.
A30.4.2
Projection microscope
A40 .
Preparation
and screen.
of equiDment.
Equipment should be readied and stored (using
A40.1 For both methods.
protective covers or other suitable enclosures) in a cleanroom or clean zone
having an airborne particulate cleanliness equal to or cleaner than that of the
Personnel performing sampling, sizing, and
cleanroom or clean zone to be tested.
counting operations should wear garments consistent with the airborne particulate
cleanliness class of the cleanroom or clean zone to be tested.
Using rinse fluid, wash the internal surfaces of all Petri slides that will be
used-to hold and transport membrane filters after the sampling and during
counting.
Allow the Petri slides to dry in a clean unidirectional airflow.
18
I
FED-sTD-209E
September 11, 1992
a
A40.2
Preparation
for the Aerosol Monitor Method.
A40.2.1
Determining background count. If an average background count for a
package of monitors (in the particle size range of interest) is provided by the
manufacturer, examine 5% of the monitors in the package and determine the average
background count on the membranes by ueing the method in A70. If the count so
.
.
obtained equals or is less than the manufacturer’s value, use the latter as the
background count for all monitors in the package. If the count so obtained is
higher than the manufacturer’s value, or if no value was provided, determine a
background count for each monitor used.
A40.2.2
Packaaina and handlina of aerosol monitors. After a background count
has been determined, place the aerosol monitors in clean containers and transport
them to the sampling locat”ion. Aerosol monitors should be opened only at the
sample location or to remove a membrane filter.
A40.3
Preparation
for the Onen Filter Holder Method.
A40.3.1 Determining backtaround count. Determine a representative background
count for the membrane filters from each box of filters to be used. Examine two
or more filter membranes per box, at 40-fold or higher magnification, using the
procedure described in A70, and record the average count.
@
A40.3.2 Cleanina the membrane filter holder and mountina the filter.
Disassemble and wash the membrane filter holder. After rinsing it with rinse
fluid, allow the holder to dry in a clean, unidirectional airflow; do not wipe
dry.. With the holder still in the unidirectional airflow, use forceps to mount
a membrane filter (grid side up) in the holder.
A40.3.3 Packaaina and transport.
Place the loaded filter holder into a clean
container and transport it to the sampling location. The holder should be
exposed only when sampling is about to take place or when removing or replacing a
membrane” filter.
A50 .
SamPlina the air.
ASO.1 Orientation and flow. When sampling air in cleanrooms and clean zones
aerosol monitor or filter holder to face
with unidirectional airflow, orient the
into the airflow and adjust the rate of sampling to achieve isokinetic conditions
(see Appendix C).
When sampling air in cleanrooms and clean zones with nonunidirectional airflow,
orient the aerosol monitor or filter holder so that the opening faces upward,
unless otherwise specified; the airflow into the filter should be adjusted to
0.00012 m’/s (0.25 ft3/min) for a 25-mm filter or 0.00047 m’/s (1 ft3/min) for a
47-mm filter.
For Class M 4.5 (Class 1000) the volume of air sampled should not be less than
0.28 m’ (10 ft’); for Class M 5.5 (Class 10 000) and classes less clean, no less
than 0.028 u? (1 ft’) of air should be sampled.
a
19
FED-STD-209E
September 11, 1992
A50.2
Usinu the Aerosol Monitor Method.
A50.2.1
Setup. At the location to be sampled, remove the bottom plug from an
Connect the monitor in series with the aerosol adapter, the
aerosol monitor.
limiting orifice or flowmeter (or orifice-flowmeter combination), and the vacuum
source. Position the aerosol monitor as required.
If a pump is used, it should either be exhausted outside the area being sampled,
or else appropriately filtered, to avoid contaminating the clean environment.
If
a flowmeter is used, adjust the flow to obtain the specified sampling flow rate.
Remove the top portion of the aerosol monitor and store it in
A50.2.2
Samplinq.
a clean location. Activate the vacuum source, start the timer, and sample the
air for a time sufficient to provide the reqyired volume of air at the selected
flow rate. When that time has elapsed, remove the aerosol monitor from the
The bottom plug need
vacuum train and replace the top portion of the monitor.
not be replaced.
Identify the aerosol monitor with a sample identification tag.
Transport the aerosol monitor to a clean zone for counting; the clean zone should
have an airborne particulate cleanliness equal to or cleaner than that of the
clean zone sampled.
A50.3
Usina the Open Filter Holder Method.
A50.3.1 Setun. At the location to be sampled, connect the filter holder in
series with the limiting orifice or flowmeter (or orifice-flowmeter combination),
and the vacuum source. Position the filter holder as required.
If a pump is used, it should either be exhausted outside the area being sampled,
or else appropriately filtered, to avoid cont~inating
the clean environment.
If
a flowmeter is used, adjust the flow to obtain the specified sampling flow rate.
A50.3.2
SamDlinq.
Remove the cover from the membrane filter holder and store it
in a clean location. Activate the vacuum source, start the timer, and sample the
air for a time sufficient to provide the required volume of air at the selected
flow rate. When that time has elapsed, remove the filter holder from the vacuum
train and replace the cover. Identify the filter holder with a sample
identification tag. Transport the filter holder to a clean zone for counting;
the clean zone should have an airborne particulate cleanliness equal to or
cleaner than that of the clean zone sampled.
A60 .
Calibration
of the microscope.
A60.1 Setup. Verify that the microscope has eyepiece-objective combinations
capable of 100- to 250-fold magnification.
Adjust the lamp and focus the
microscope to illuminate evenly the entire field of view. Place the stage
micrometer on the mechanical stage. Adjust and focus each eyepiece independently
to give a sharp image of the gradations on the stage micrometer.
20
●
FED-STD-209E
September 11, 1992
I
e
If an image analyzer or projection microscope is used, perform a similar
calibration.
I
.
A60.2 Procedure.
The following steps axe used to calibrate a specific ocular
reticle paired with a specific stage micrometer for the measurement of particles
at any selected level of magnification.
(a) Determine and record the number of stage micrometer divisions, S, of
size M (micrometers)* corresponding to the number of divisions, R, in the
full scale of the ocular reticle for each magnification of interest.
(b) Calibrate the scale of the ocular reticle for a given
using the formula:
SxM/R=
magnification
Micrometers per scale division of the
ocular reticle
(Equation A60-1)
Example:
For a given ocular reticle and stage micrometer at 100-fold
magnification, let 150 divisions of the reticle correspond to 100
divisions, each 5.0 pm in length, of the stage micrometer.
Using
equation A60-1,
SxM/R=
(100 divisions) x (5.0 pm/division)/(150 divisions)
= 3.33 pm per scale division of the ocular reticle
(c) Calculate the number of divisions of the ocular reticle
corresponding to each specific particle size of interest.
Example:
Using the same data as in (b), calculate the number of divisions of the
ocular reticle required to size particles in the range of 10 to 2P ~.
Since, at 100-fold magnification, each division of the ocular scale
equals 3.33 pm, counting particles whose longest dimensions span 3 to 6
divisions will size particles in the range of 10 to 20 pm.
If the microscope has a zoom mechanism, appropriate intermediate magnifications
may be selected in order to calibrate the ocular scale to integral values only.
A change in interpupillary distance between operators changes focal length and,
therefore, calibration.
21
FED-STD-209E
September 11, 1992
A70.
Cotmtina and sizinq Darticles by optical microscoDv.
A70.1 Setup. In a cleanroom or clean zone suitable for the counting and sizing
of particles, remove the membrane filter from the aerosol monitor or open filter
holder using forceps.
Insert the membrane, grid side up, in a clean Petri slide
and cover it with the lid. Place the Petri slide on the microscope stage.
Adjust the angle and focus of the illuminator to provide optimum particle
definition at the magnification used for counting. Use an oblique lighting angle
of 10 to 20 degrees so that the particle casts a shadow, thus enhancing
definition.
A70.2
Selectina a field size. Select a field size which contains fewer than
approximately 50 particles, “5 pm and larger. Possible choices are: a single grid
square, a rectangle defined by one side of a grid square and the entire
calibrated scale in the ocular reticle, or a rectangle defined by one side of a
grid square and a portion of the calibrated scale of the ocular reticle.
A70.3 Countina Darticles.
Estimate the total number of particles, 5 pm and
larger, present on the membrane filter by examining one or two of the selected
fields. If this estimate is greater than 500, use the procedure for counting
particles described in A70.4.
If the estimate is less than 500, count all of the particles on the entire
effective filtering area of the membrane.
Scan the membrane by manipulating the
stage so that the particles pass under the calibrated ocular scale. The size of
The eyepiece with its
a particle is determined by its longest dimension.
calibrated ocular scale may be rotated if necessary.
Using a manual counter,
tally all particles with sizes in the range of interest. Record the riumber of
particles counted in each field.
A70.4 Statistical particle countinq. When the estimate of the number of
particles, 5 pm and larger, on the membrane filter exceeds 500, a statistical
counting method should be used. After a unit field size has been selected,
particles are counted in a number of fields of that size until the following
statistical requirement is met:
(Equation A70-1)
FXN>500
where:
F
N=
= number of unit fields counted, and
total number of particles counted in F unit field?.
22
●
FED-STD-209E
September 11, 1992
The total number of particles on the membrane is then calculated
following equation:
from the
(Equation A70-2)
P =NxA/(Fxa)
where:
P=
total number of particles in a given size range on the membrane,
N = total number of particles counted in F unit fields,
F=
number of unit fields counted,
a = area of one unit field, and
A=
total effective filtering area of the membrane.
A80 . ReXrtinq.
Subtract the background count from the total number of
Calculate the airborne particulate concentration of
particles on the membrane.
the air sampled by dividing the number of particles collected by the sample
volume. Results may be expressed for each size range of interest.
e
A90. Factors affectina nrecision and accuracy. The precision and accuracy of
this method are subject to human and mechanical error. To minimize human error,
technicians must be trained in microscopy and in the sizing and counting of
particles.
Experienced technicians are also more likely to note deficiencies in
equipment, further reducing the possibility of error. Standard specimens may be
obtained or prepared for use in training technicians in the counting and sizing
of particles.
For a given location, the repeatability of this method can be improved by
increasing the number of samples or increasing the volume of air sampled, or
both .
23
FED-sTD-209E
September 11, 1992
APPENDIX B
OPERATION OF A DISCRETE-PARTICLE
B1O .
COUNTER
Scope and limitations.
B1O.1 ScoDe. This appendix describes methods for the testing and operation of
discrete-particle counters (DPC’S) used to satisfy the requirements of this
Standard.
DPC’S provide data on the concentration and size distribution of
airborne particles within the approximate range 0.01 to 10 pm on a near-real-time
basis. A DPC will correctly size only those particles within the limits of its
dynamic range. The optical particle counter and the condensation nucleus counter
are representative of single particle counting instruments.
..
B1O;2 Limitations.
Data related to the size and size distribution of particles,
obtained through the primary calibration of a DPC, are dependent upon the type of
particles used for calibration and upon the design of the DPC’S.
Care must be exercised when comparing data from samples containing particles that
vary significantly in composition or shape from the particles used for
calibration.
Differences in the design of DPC’S which can lead to differences in counting
include dissimilar optical and electronic systems, predetection sample processing
systems, and sample handling systems.
●
Potential causes of difference such as the foregoing should be recognized and
minimized by using a standard primary calibration method and by minimizing the
variability of sample acquisition procedures for instruments of the same type.
In view of the significance of these effects, ”a,detailed description of each DPC
in use should be recorded.
B1O.3 Qualifications of Dersonnel.
Individuals supervising or performing the
procedures described herein should be trained in the use of DPC’S and should
understand the operation, capabilities, and limitations of the instruments.
B20.
References.
B20.1 ASTM F50, Standard’Practice for Continuous Sizing and Counting of Airborne
Particles in Dust-Controlled Areas and Clean Rooms Using Instruments Capable of
Detecting Single Sub-Micrometer’ and Larger Particles.
B20.2 ASTM F328, Practice for Determining Counting and Sizing Accuracy of an
Airborne Particle Counter Using Near-Monodisperse Spherical Particulate
Materials.
B20.3 ASTM F649, Practice for Secondary Calibration of Airborne Particle Counter
Using Comparison Procedures.
24
,.
FED-sTD-209E
September 11, 1992
a
B20.4 IES-RP-CC013, Recommended Practice for Equipment Calibration or Validation
Procedures, Institute of.Environmental Sciences.
B20.5 Scheibel, H. G., and Porstendorfer, J., “Generation of Monodiaperse
Ag- and NaCl Aerosols with Particle Dkseters between 2 and 300 rim,” J. Aerosol
Sci., M(2),
113-125 (1983).
B20.6 Bartz, H., et al, “A New Generator for Ultrafine Aerosols below 10 rim,”
Aerosol Sci. Technol., Q(2), 163-171 (1987). ‘
B20.7 Keady, P. B., and Nelson, P. A., “Monodisperse Particle Generators for
Calibrating Aerosol Instrumentation,” Proc. IES Ann. Tech. Mtg., Orlando,
Florida, May 1, 1984.
B20.8 Li.u, B. Y. H., Pui, D. Y. H., Rubow, K. L., and Szymanski, W. W.,
“Electrostatic Effects in Aerosol Sampling and Filtration,” Ann. Occup. Hyg.,
~(2), 251-269 (1985).
B20.9 Raasch, J., and H. Umhauer, “Errors in the Determination of Particle Size
Distribution Caused by Coincidence in Optical Particle Counters,” Particle
Characterization, ~(l), 53-58 (1984).
B2O.1O
Nz, h,
@
Niida, T., et al, “Counting Efficiency of Condensation Nuclei Counters in
1417-1420 (1988).
C% and He,” J. Aerosol Sci., E(7),
B20.11 Gebhart, J., and Roth, C., “Background Noise and Counting
Efficiency of Single Optical Particle Counterst” Aerosols: Formation and
Reactivity (Proc. Second International Aerosol Conference, West Berlin, Germany,
1986), Pergamon Journals, Ltd., Oxford, England (1986) 607-611.
B20.12 Ramey, T. C., “Measuring Air Flow Electronically,”
Engineering, ~(5),
29-33 (1986).
Mechanical
B20.13 Baker, W. C., and Pouchot, J. F., “The Measurement of Gas Flow,
Part 1,” J. Air Pollution Control Association, ~(l),
1983.
“The Measurement of Gas Flow,
B20.14 Baker, W. C., and Pouchot, J. F.,
Part 11,” J. Air Pollution Control Association, =(2),
1983.
B30.
Summarv of method.
.
B30.1 S~cifvinu
a Brocedure.
A sample acquisition procedure should be
established based on the level of cleanliness of the air that is to be verified
or monitored.
This program should include a description of the DPC or DPC’S to
be used, the sample transport system, the inlet probe, and any other features
related to the operation of the DPC. The range of particle sizes to be measured
should be identified as well as the sample volume and the location and frequency
of sampling.
If measurement is required over a very wide range of particle
sizes, then more than one DPC may be required. The range for accurate
measurement of pmicle
size (dynamic range) by a DPC will vary with
sensitivity.
For a DPC used only to size particles smaller than 1 pm, a dynamic
a
25
FED-STD-209E
September 11, 1992
range of 20:1 is typical.
For a DPC used to size particles larger than 1 pm, a
dynamic range of up to 40:l,is typical.
The dynamic range of a DPC depends upon
the particle size distribution being measured and the gain of the data processing
system.
●
B30.2 Calibration.
Calibration of the DPC is required for the counting and
sizing of particles and to verify the sample flow rate. Size calibration is
performed with isotropic particles.
Calibration for concentration is carried out
with either monodisperse or polydisperse particles, as described in recognized
standard methods (see, for example, B20.2 and B20.3). Latex spheres of well
defined or certified mean diameter and standard deviation can be used to
calibrate DPC’S for particle size definition.
Alternatively, calibration
particles, can be produced by physically separating a sized fraction of particles
from a polydisperse suspension.
The fraction may be defined either at the lower
size limit or at both the upper and lower limits. The fractionating device
should be defined and the size of the calibration particles stated with reference
to the process used for fractionation.
Monodisperse particles may also be
produced by controlled condensation from a vapor (see, for example, 2.5 and 2.6)
or by controlled atomization from a vibrating orifice (see 2.7). When
calibration particles are produced by either of these methods from a material
with a refractive index different from that of latex particles, it is important
to note that the DPC being calibrated may indicate different particle sizes for
the different materials, even though the particles are the same size.
Stable operation of the DPC can be achieved by standardizing against internal
references built into the DPC or by other approved methods (see, for example,
B20.1 through B20.4).
B30.3 ODeration.
Air in the cleanroom or clean zone to be verified or monitored
is sampled at a known flow rate from the sample point or points of concern.
Particles in the sampled air pass through the sensing zone of the DPC. Each
particle produces a signal that can be related to its sizer either directly or
.“
with reference to the operation of a predetection sample processing system. An
electronic system sorts and counts the pulses, registering the number of
particles of various sizes which have been recorded within the known volume of
air sampled. The concentration and particle size data can be displayed, printed,
or further processed locally or remotely.
B40 .
Apparatus and related documentation.
B40.1. Particle countinu system. The apparatus should consist of a DPC selected
on the basis of its ability to count and size single particles in the required
size range. The DPC should include a sample airflow system, a particle sensing
and measuring system, and a data processing system. The particle sensing and
measuring system may include a means of size fractionation prior to particle
sensing and measuring.
The sensitivity of the DPC (minimum measurable particle
size) should be selected consistent with the requirement for verifying that the
air complies with the airborne particulate cleanliness class in the area of
interest.
For verification based on the measurement of particles approximately
0.1 pm and larger, an optical particle counter, a time-of-flight particle sizer,
or an.equivalent counter can be used. For verification based on the measurement
26
:.
●
FED-STD-209E
September 11, 1992
of ultrafine particles, a counter such as a condensation nucleus counter, alone
or in combination with a ,diffusion battery, a differential mobility analyzer, or
an equivalent device can be used.
B40 .2 Sanmle airflow svstem. The sample airflow system consists of a sampling
probe with a sharp-edged inlet, a transit tube, a particle sensing and measuring
chamber, an airflow metering or control system, and an exhaust system. No abrupt
transitions in dimension should occur within the airflow system. The probe is
which transports the sampled air to the patticle
connected to a transit tube
sensing chamber.
Probes that approach isokinetic sampling conditions can aid in
reducing sampling bias (see Appendix C). The tube should have dimensions such
that the transit time in the tube does not exceed 10 seconds.
,.
The probe and transit tube should be
B40.2.1 Particle transit considerations.
configured so that the Reynolds number is between 5 000 and 25 000.
For particles in the range of 0.1 to 1 pm and for a flow rate of 0.028 #/rein
(1.0 ft3/min), a transit tube up to 30 m long may be used. For particles in the
range of 2 to 10 pm the transit tube should be no longer than 3 m. Under these
conditions, losses of small particles by diffusion and of large particles by
sedimentation and tipaction are predicted to be no more than 5% during transit
through the tube (see Appendix C). For most applications, these tube
For special situations,
configurations and flow conditions will be satisfactory.
more precise particle transit characteristics can be calculated (see B20.8).
a
B40.2.2 Flow control and exhaust air filtration. The sample airflow system
should contain a flow induction device and a means of metering and controlling
the flow. The flow induction device may be either a built-in or an external
vacuum source. The system for metering and controlling the flow of sample air
should be located after the particle sensing chamber in order to minimize
particle losses and the generation of artifacts before sensing has taken place.
If a built-in vacuum pump is used, the air exhausted from the pump should be
suitably filtered or vented to prevent particles in the sampled air stream, as
well as those generated by the pump, from being exhausted into the controlled
environment.
In addition, particles may emanate from the interior of the DPC,
for example from a cooling fan or by other movement of air through the counter.
Such particle-laden air must be suitably filtered or vented to prevent it from
contaminating the air being sampled as well as the clean zone in which the DPC
is operating.
B40.3 Sensina and measurina chamber. The sensing system of the DPC is limited
in volume so that the probability of more than one particle being present at any
time (coincidence error) is less than 10% (see B20.9). The operation of the
particle sensing chamber will be defined by the nature of the DPC. Since
uncontained sample flow may occur within the chamber, its design should be such
that minimum recirculation and recounting of particles occur in that chamber.
If
the particle characterization system includes any particle manipulation
(e.g., ”diffusion battery, electrostatic charging system, or nucleation chamber)
before sensing occurs, then the DPC element used to control or limit the size of
the particles counted should be such that no significant undefined change in the
The detection
number of countable particles takes place during that process.
@
27
FED-STD-209E
Septeher
11, 1992.
elements within the sensing chamber should be designed to maintain stated
accuracy, despite normal variation in specified operating line voltage and
ambient temperature.
B40.4 Electronic system. The data processing system of the DPC should include
components for counting and sizing (or merely counting) signals from the
particles observed by the DPC, a means of converting signal levels to particle
sizes, sufficient data processing capability to convert the number of particles
counted and the volume of air sampled to particle concentration, and internal
monitoring capability to verify that critical DPC components are operating
correctly.
Data should be available as front-panel display, on-board hard copy,
or as signals that can be transmitted to a remote data reception device in a
The
format that will allow either direct storage or further processing.
processing system should also include the necessary components to carry out
or
standardization of the DPC. The standardization may be done eithermanually
automatically.
B40.5 Standardization.
An internal standardization or secondary calibration
system or other means of ensuring stability should be provided in the DPC. The
standardization system should be capable of validating the stability of the DPC’S
operating parameters.
The secondary calibration system is used to check the
stability of the counting and sizing capability of the DPC and to provide a
stable reference for any necessary adjustments in sensitivity.
B40.6 Documentation.
manufacturer include:
(a)
(b)
(c)
(d)
(e)
(f)
(9)
(h)
(i)
(j)
Instructions which should be supplied with the DPC by the
Brief description of the DPC’S operating principles
Description of major components
Environmental conditions (ambient temperature, relative humidity, and
pressure) and line voltage range required for stable operation
Size and concentration range of particles for which measurements are
accurate
Suggested maintenance procedures and recommended intervals for routine
maintenance
Operating procedure for counting and sizing particles
Secondary calibration procedure (where applicable)
Procedure and recommended interval for primary calibration, as well as
provision for calibration by manufacturer upon request
“
Field calibration procedures and capability
Recommended supply and estimated usage,of consumable items
B50. Preparations for sanmlinq. The procedures described in the following
paragraphs should be performed before using a DPC to verify the compliance of
air to an airborne particulate cleanliness class. Each DPC has its own
requirements with respect to the frequency for performing these procedures.
B50.1 Primary calibration.
Primary calibration of a DPC entails characterizing
its ability to size and count airborne particles with known accuracy in a
measured volume of air. The following paragraphs contain guidelines to be
considered when using calibration procedures for DPC’S described in the
It may be
literature (see, for.ex~ple,
B20.2, B20.3, B20.4, and B2O.1O).
necessary to deviate from the methods in such documents to achieve a specific
28
FED-STD-209E
September 11, 1992
e
objective.
For a DPC that includes a pre-counting particle size fractionation
system (such as a diffusion screen or a system that responds to electrostatic
charge), operation of such a system may also require calibration.
B50.1.1 Particle sizinq. Pri.marycalibration
of the particle sizing function of
the DPC is carried out by registering its response to a monodisperse, homogeneous
aerosol (containing predominantly spherical particles of known size and physical
properties), and by setting the calibration control function so that the correct
size is indicated.
Thereafter, the internal secondary calibration system is
adjusted, if necessary, to maintain a stable response to a reference aerosol
suspension.
Nonspherical particles may be used for primary calibration in
specific applications.
The particle size is then defined in terms of an
appropriate dimension for the reference particles.
Means of generating reference
particles have been extensively described in the literature.
B50.1.2 Particle countinu efficiency.
The counting efficiency of a DPC is
affected by a number of operating characteristics.
For smaller particles,
instruntent sensitivity and background noise are important, and procedures for
defining the counting efficiency of a DPC for such particles are discussed
in B20.2.
a
For particles larger than approximately 5 pm, counting efficiency is also
affected by the DPC’S sampling efficiency and by transport effects.
The counting
efficiency of a DPC for such larger particles can be determined by means of a
referee method.
The referee method may be a sampling and measurement system
which is identical (or not) to that of the DPC being tested. The procedure
(see B20.11) consists of generating within a chamber an aerosol composed of large
particles, drawing a sample of that aerosol into both the DPC and the referee
measurement system, and determining the ratio of particles counted by the DPC and
the referee system.
B50.1.3 Air sasmle volume. The air sample volume is calibrated by measuring the
flow rate and the duration of the sampling interval (see B20.12, B20.13, and
B20.14).
If the DPC measures only those particles within a specified portion of
the air sampled by the airflow system, information must be obtained from the
manufacturer in order to calibrate both the inlet air sample volume and the air
volume in which particles are measured.
To avoid errors, equipment used for
these measurements should not introduce an additional static pressure drop. All
flow measurements should be referenced to ambient conditions of temperature and
pressure or as otherwise specified.
.
a
B50.2 False count or background noise check. A check for false counts or the
measurement of background noise, or both, should be performed in the cleanroom to
be characterized.
A filter capable of removing at least 99.97% of particles”
“ equal to and larger than the size of smallest particle detectable by the DPC is
connected to the inlet. After adjusting the flow to the correct rate, the count
rate is recorded for the smallest particle detectable by the DPC. The DPC should
record no more than an average of one false count during the measuring period
required to collect the minimum sample vo~ume indicated in 5.1.3.4.
If more than
one count per”period is registered in that size range, the DPC should be purged
with the filter in place until an acceptable level of false coun”ts is achieved.
29
FED-STD-209E
September 11, 1992
B50.3 Field [secondarv) calibration Procedures.
Standardize the DPC in
accordance with the manufacturer’s instructions.
The count rate from background
noise, recorded at the time of primary calibration, should also be checked during
field calibration.
B60 . Samulinq.
Perform the background noise check and the field calibration in
accordance with B50.2 and B50.3. Check the sample flow rate and adjust it to the
specified value, if applicable.
Turn on the counting circuits and data
processing components, if necessary.
Collect data for the particle size(s) of
concern.
B60.1 SamPlinu for verification.
When sampling air for the purpose of verifying
its compliance to an airborne particulate cleanliness class, sufficient data
should be obtained to satisfy the statistical requirements of 5.4. Sample
locations should be established in accordance with 5.1.3. Proper orientation of
the probe should be established in accordance with 5.3. Recommendations for
proper lengths of sample transit tubes should be observed in accordance with
B40.2.1.
B60.2 SamDlina for monitoring.
Sampling procedures should be established in
support of the monitoring plan described in 5.2.1. Sample locations may be
established in accordance with 5.1.3, or as appropriate.
Probe orientations may
be established in accordance with 5.3, or as deemed appropriate for specific
monitoring situations.
Sampling in support of monitoring need not meet the rigid
statistical criteria required for verification; the observation of’trends and
anomalies, without application of rigorous statistical limitations, is generally
more appropriate.
For monitoring purposes, the lengths of sample transit tubes
may deviate from those recommended in B40.2.1.
Record the following information, as specified, for the
B70. Renortinq.
verification of air in a cleanroom or.clean zone to an airborne particulate
cleanliness class, or for the monitoring of air cleanliness:
(a)
(b)
(c)
(d)
(e)
(f)
(9)
(h)
(i)
(j)
(k)
(1)
Identification and location of the cleanroom (or clean zone)
Identification of the DPC and its calibration status
Background noise count for the DPC
Date and time when the DPC was used
Cleanroom (or clean zone) status: “As-built,” “At-rest,”
“Operational ,“ or as otherwise specified
Type of test, verification or monitoring
Target” level of verification of the cleanroom or clean zone
Range(s) of particle sizes measured
DPC inlet sample flow and sensor measured sample flow
Location of sampling points
Sampling schedule for verification or sampling protocol for monitoring
Raw data for each sample .point, as required
30
FED-sTD-20,9E
September 11, 1992
APPENDIX C
ISOKINETIC AND ANISOKINETIC
SAMPLING
Clo. -.
This appendix presents formulas for determining whether isokinetic
sampling conditions exist and, if they do not, for estimating the artificial
change in concentration caused by anisokinetic ssmpling.
C20.
Reference.
Czo.1
Hinds, W. C., Aerosol Technolo qv: Properties, Behavior, and Measurement of
Airborne Particles, John Wiley & Sons, New York (1982), 187-194.
C30. Background.
When particles are sampled from a flowing air stream, a
difference between the air velocity, in the stream and the air velocity entering
the probe inlet can cause a change in concentration because of particle inertia.
When these velocities are the same, the sampling is isokinetic; otherwise, the
sampling is anisokinetic.
Isokine~ic sampling is achieved when the probe inlet is pointed into the
direction from which the flow is coming and is parallel with (isoaxial to) that
flow, and when the mean flow velocity into the inlet matches the mean flow
velocity of the air at that location.
e
C40. 14ethods. The mean velocity of the air in the probe inlet is v = Q/A, where
v is the velocity, Q is the volumetric rate of airflow into the inlet, and A is
the cross-sectional area of the inlet.
Figures C.1 and C.2 show the diameters of circular inlets of probes that will
produce isokinetic sampling at the indicated air velocities and volumetric flow
rates. If the velocities are matched to within 5% of one another, isokinetic
conditions are considered to exist and no correction is needed to determine the
concentration of airborne particles.
If isokinetic conditions cannot be achieved, formulas are available for
,
predicting the concentration in the sample, C, in terms of the concentration in
the flowing air, CO, the mean sampling velocity, v, and the free-stream velocity
of the air, VO. The formula of Belyaev and Levin as presented in c20.1 applies:
c/c. = 1 + (vo/v - 1)”{1 - 1/[1 + {2 + 0.62-(v/v.)}*Stkl)
(Equation C40-1)
In this equation, the Stokes parameter,
(Equation C40-2)
Stk = -0/D,~
appears, which is based on the particle aerodynamic relaxation time, 7, the
free-stream velocity of the air, VO, and the inlet diameter of the probe, D,.
31
FED-STD-209E
September 11, 1992
The aerodynamic relaxation time for a spherical particle is:
‘r= C=PdP2/18q,
(Equation C40-3)
where C=, the Cunningham correction, is expressed as
cc = 1 + 0.16 x 104 cm/dP,
(Equation C40-4)
and
dP = the particle diameter
(cm),
P = the density of the particle
(g/en?), and
11 = the viscosity of air (1.81 x 104 poise at 20 “C).
Therefore, for particles with the density of water and a diameter, d, at room
temperature and pressure, the aerodynamic relaxation times are (see C20.1):
d
(Pm)
(:)
0.1
0.2
8.85 X 104,
2.30 X 10-T
0.3
4.32 X
0.5
5.0
1.02 x 104
7.91 x 10*
10-7
For Stk >> 1 (large particles, fast flows, probes with small inlet diameter
inertial effects predominate and C/CO becomes vO/v. For Stk << 1, inertial
effects are negligible, and C/CO is nearly equal to 1.0. Between these limi
calculations can be made to gauge the importance of mismatches in velocities.
Where one is free to chooge Q or D, or both, the velocities can be matched by
setting:
Q/A = Q/(7r/4).D,2= V.
“(Equation C40-5)
Oftenr with Q fixed and with limited choices for the diameter of the probe inlet,
this degree of flexibility is not available.
In such cases, select D, as close
to the optimum as possible, then determine if C/CO is close enough to unity such
that no further correction is needed.
32
1
FED-STD-209E
September 11,
,!IB
21
I
moo
I
crn3J~ ,
I
I
I
I
I
II
4
56
I
I
I
I
I H
019
8
.
7
6 5
4
1.
2 ~
0°9
I
8
6
5
3
o’
i
AIR VELOCITY,
7
8 9102
V. (cm/S)
Figure C. 1. Probe inlet diameters (metric units) for isokinetic
sampling, v = VO (volumetric flow rates shown parametrically)
019
8
7
L%
3
0°
‘-l-
1
!
I
87m
a
“2
3
4
56
7
AIR VELOCllY
8
’102
2
VO (ft/mifO
Figure C.2. Probe inlet diameters (English units) for isokinetic
sampling, v = vO (volumetric flow rates shown parametrically)
33
FED-sTD-209E
September 11, 1992
In Figure C.3 the velocity ratios, v/vO, are plotted vs. Stk for those conditions
which result in sampling bias contours for C/CO that are within ~5% of 1.0, i.e.,
0.95 and 1.05. The smaller the particle, the smaller T, thus the smaller Stk,
and the wider the range of acceptable velocities.
o
Two particle diameters are of particular interest: 0.5 pm and 5 pm. It can be
shown that anisokinetic sampling is unlikely to have a significant effect on
particles 0.5 pm and smaller.
If the air is being sakpled for the total
concentration of particles 0.5 pm and larger, typically the count will be
dominated by particles with diameters near 0..5pm; if these particles are not
much affected by anisokinetic conditions, neither will the total count be
affected.
Thus, anisokinetic sampling in clean zones is likely to be significant
only when sampling at 5 pm and larger.
.
;.
0
s 10’=z?
G
25
w
\
\
6
\
4
>
IN
CORRECTIC
\
Z3
REQUIRED
c/co<o.95
$
m
G
~
2
NO
CORRECTI{
>N
REQUIRED
1oo~
z.
G
7
96
m
s
w
>
4
a
z
3
CORRECTiC
)N OPTIONAL
c/c~>l.05
3
n
Z2
$
L
o 10-1
10-3
0
i=
~
10-2
STOKES
10-1
Stk =
PARAMETER,
1O’J
TVo/DS
Figure C.3. Contours of sampling bias, C/CO = 0.95, 1.05
In a clean zone with unidirectional flow, VO is typically 50 cm/s (100 feet per
minute) or less. Under such conditions, values of Stk for probes of several
selected inlet diameters, D,, and two particle diameters, d, are as follows:
D,
(cm)
0.1
0.2
0.5
1.0
2.5
Stk
(d = 0.5 pm)
0.00051
0.00026
.0.00010
0.00005
0.00002
34
(d = 5 pm)
0.040
0.020
0.0080
0.0040
0.0016
FED-sTD-209E
September 11, 1992
a
Under theee conditions, velocity ratios as extreme a~ v/vO = 10 and v/vO = 0.1
will not cause so much ae a 5% sampling error for 0.5-Laa particles.
Under these
same conditions, for 5-pm particleg, a-probe with an inlet diameter greater than
0.5 cm would allow velocity ratios between about 0.3:1 and 7:1 without giving a
predicted error greater than 5%, but a probe with an inlet diameter as small as
0.1 cm would allow velocity ratios only between about 0.7:1.0 and 1.8:1 for 5% or
less sampling error.
The analysis indicates that anisokinetic sampling of particles 0.5 pm and smaller
is not a problem in typical clean zones and will rarely be a problem when
sampling at 5 pm unless sampling is carried out to detect a point source of
particles.
If one “is sampling particles S pm and larger in a clean zone and if the mismatch
in velocities is greater than 5%, then the equation for C/CO (or Figure C.3)
Sf the correction
should be used to calculate the magnitude of the correction.
is greater than 5%, then it should be applied to the observed concentration if
‘doing so raises the observed concentration, as in the case where the sampling
The correction may also be
velocity is more than the air free-stream velocity.
applied to the observed concentration if doing so decreases the observed
concentration, as in the case where the sampling velocity is less than the air
free-stream velocity.
C50. Examle.
Assume one plans to sample particles 5 P and larger in a flow
that has a velocityf VO = 1 m/s (100 cm/s) using a probe with an inlet diameter
of 1 cm.
From the table, the aerodynamic relaxation time for a particle with a S-pm
aerodynamic diameter is 7.91 x 10a s. The Stokes parameter, then, becomes
Stk = 0.00791.
10 cm/s, then the volumetric
The mean sample flow velocity is v = Q/A. Ifv=
flow rate through the probe is 7.85 cn?/s (since Q = v“A), and-v/vO = 0.1. Thus
an artificially high concentration will be produced:
c/c.=1+9*(1-
1/[1 + (2.062)(0.00791)])
= 1.144
In this example, the anisokinetic bias in the sample of air is corrected by
dividing the observed concentration of particles, C, by 1.144 to obtain CO, the
concentration of particles in the flowing air stream.
35
FED-sTD-209E
September 11, 1992
APPENDIX D
METHOD FOR MEASURING THE CONCENTRATION OF ULTRAFINE PARTICLES
/
D1O. -.
This appendix describes procedures for measuring the concentration
of ultrafine airborne particles, particles larger than approximately 0.02 pm, for
comparison with the specified U descriptor.
It also defines the cutoff
characteristic required for discrete-particle counters’ (DPC’S) to be used to
measure the U descriptor.
D20.
References.
D20.1 Cheng, Y. ,S., and Yeh, H. C., “Theory of a Screen-Type Diffusion Battery,”
J. Aerosol Science, ~, 313-320 (1980).
D20.2 Cheng, Y. S., and Yeh, H. C., Aerosols in the Minina and Industrial Work
Environmentsr Marple, V. A., and Liu, B. Y. H., Eds.; Ann Arbor “Science
Publishers, 1077-1094 (1983).
D20.3 Hinds, W. C., Aerosol Technolouv : Properties, Behavior, and Measurement
Airborne Particles, John Wiley & Sons, New York (1982).
of
D20.4 Liu, B. Y. H., and Pui, D. Y. H.’, “A Submicron Aerosol. Standard and the
Primary, Absolute Calibration of the Condensation Nuclei Counter,”’ J. Colloid and
Interface Science, 47(l), 155-171 (1974).
To verify the U descriptor, a condensation nucleus counter
D30. Amaratus.
(or other DPC as described in Appendix B) having a dynamic size range of at least
0.02 to 1.0 pm shall be used. Only a DPC whose counting efficiency curve meets
the criteria specified in D30.1 should be used.
e“
D30.1 Countina efficiency.
The counting efficiency characteristic of the DPC
used to verify a U descriptor must fall within the shaded envelope of Figure D.1.
This region of acceptable performance centers on a counting efficiency of 50% at
0.02 pm and includes a tolerance band of 0.002 pm on either side of 0.02 #m. The
minimum and maximum counting efficiencies that are acceptable outside the
0.018-to-O.022-pm tolerance band are based on the calculated penetration Of.a
diffusion element (see D20.1) having either 40% efficiency at 0.02 pm (the branch
for particles with diameters larger than 0.022 pm) or 60% efficiency at 0.02 pm
(the branch for particles with diameters smaller than 0.018 pm).
The counting efficiency curve of a DPC can be determined using the methods of
Liu and Pui (see D20.4). Manufacturers of DPC’S will normally provide this
information on request.
If the DPChas a counting efficiency curve that falls to the right of the
envelope in Figure D.1, the DPC cannot be used to verify the U descriptor.
If
the curve falls to the left of the envelope in Figure D.1, then the DPC may be
used to verify the U descriptor if it is subsequently modified with a size cutoff
inlet device as described in D30.2, in which case the over,all counting efficiency
of the modified DPC becomes the product of the counting efficiency of the
unmodified DPC and the fractional penetration of the cutoff inlet.
‘a
36
FED-STD-209E
September 11, 1992
I
I
a
80
60
_(O.02Wrn,50%)
;
40
J..
20
0
0.01
0.0180.022
0.03
0.04
0.05 0.06
0.08
0.1
Particle Diameter (pm)
Figure D.1.
Envelope of acceptability for the counting efficiency
of a DPC used to verify the U descriptor.
D30.2 Size cutoff inlet device. To achieve the 0.02-ffm cutoff characteristic
required to verify conformance of air to a U descriptor, a size cutoff device can
be placed on the sampling probe of a DPC whose size detection efficiency curve
falls to the left of the acceptable shaded envelope of Figure D.1. The counting
efficiency is thus reduced in this region so that the overall cutoff
characteristic for the combined instrument, sampling probe, and inlet device
falls within the shaded envelope. Cutoff devices remove small particles by
diffusional capture in a well-defined and reproducible manner.
The required penetration characteristic is achieved through diffusional capture
of small particles by means of a tube, parallel plate, or fine-mesh screen.
Other possible devices include those with collimated hole structures, packed
beds, and porous carbon disks. A wide variety of sizes and configurations of
inlet devices is possible and acceptable, providing they produce the required
penetration characteristics.
Cutoff devices are commercially availabLe; one type
is based upon the equations of Cheng and Yeh (see D20.1).
.
The penetration of paxticles through cutoff devices is a function of the
volumetric flow rate. The device should not be used at flow rates different from
To avoid accumulation of static charge, the
those for which it was designed.
device should be made of electrically conductive material and grounded.
With the cutoff
D40 . Determining the concentration of ultrafime Darticles.
device (if needed) fitted to the probe of the DPC as described above, sample the
air in accordance with section 5 of this Standard. Divide the total number of
particles by the volume of air sampled. Report the concentration in particles
per cubic meter.
37
FED-STD-209E
September 11, 1992
APPENDIX E
RATIONALE FOR THE STATISTICAL RULES USED IN FED-STD-209E
E1O. -.
This discussion surveys the two statistical rules (see 5.4.1)
embedded in this Standard, describes what the rules accomplish in practice, and
summarizes the rationale for each.
E20. The statistical rules. The first rule requires that, for each location
sampled, the average particle concentration not exceed’the class limit or U
descriptor.
The second rule requires “that an upper 95% confidence limit,
constructed from all of the location averages; not exceed the class limit or U
descriptor; this rule applies only when fewer than ten locations are sampled.
The rationale for the first rule is that the cleanliness of the air must be
checked at multiple locations in a cleanroom or clean area. The number of
locations is a function of the size of the area to be checked.
The average of
all measurements taken at a given location was selected as the statistical unit.
This unit was chosen from the perspective that FED-STD-209E is targeted at an
average level of performance rather than at an absolute one. Thus the average
particle concentration at each sampling location is the base level of statistical
summarization found in the Standard.
All variation among the measurements taken at each location is ignored, except
for the extent to which it affects the average at that location and the two
statistical rules.
It may be possible, therefore, to average out sampling
variability, spikes, and time trends in data collected at a given location.
Minimizing the impact of sampling variability is a desirable consequence of using
an average. The potential for obscuring real changes in particle concentration
(spikes, cyclicity, or other time-related trends) is an undesirable consequence
of using an average.
*
The Standard also specifies a minimum sampling volume for measurements taken at
any location.
This serves to ensure that (1) sufficient data are collected to
allow a reasonable evaluation of the cleanliness of the air in accordance with
the intent of the Standard, and (2) a sufficiently large volume is sampled to
allow the approximate application of normal distribution theory to the
measurements collected at a given location. The second condition occurs only
when actual particle concentrations are near or above the class limit:’-orU
descriptor.
This does not presents
problem when a location has a very low
concentration, compared to the specified limit, and statistical analysis of the
measurements at that location is not critical.
Requiring a minimum sampling volume does not average out all sampling
variability.
It is likely, therefore, that if the average air cleanliness is
near the class limit or U descriptor and if one or a few measurements are made at
each location, that some locations will be found which exceed the specified
limit. The only statistical fix for such locations is to collect additional
measurements at those locations. Thus, the Standard @plicitly
requires
more,
sometimes many more, measurements to be made when average performance is near the
specified limit.
38
:
FED-sTD-209E
September 11, 1992
The rule requiring that the average particle concentration at each location meet
the class limit or U descriptor is the dominant statistical rule for all but the
smallest facilities.
Because this rule implicitly ignores the effect of sampling
variability, it becomes increasingly difficult to obtain a passing result as more
locations are required to be sampled. Thus, the number of locations sampled will
affect the probability of “passing the Standard. This rule also has the
undesirable effect of discouraging sampling at any more than the minimal number
of locations required by the Standard whenever the number of locations equals or
exceeds five.
The highest level of statistical summarization used in the Standard is the 95%
upper confidence limit placed on the grand average of all the averages obtained
at each of the locations sampled. All variation among the location averages is
ignored other than how it impacts the calculation of the upper confidence limit.
It may be possible, therefore, to average out real differences in average
cleanliness from location to location or differences observed over time so long
as the average at each location stays below the specified limit.
Construction of the upper confidence limit is based on use of a t-table which is
provided (Table II) as part of the Standard. This rule implicitly assumes that
the distribution of the averages at each location originates from the same normal
distribution or that sufficient locations have been sampled for the central ltiit
theorem to be invoked. Sometimes neither of these implicit assumptions will be
met. However, the UCL based on the t-statistic is reasonably robust even to
moderate violations of these implicit assumptions.
The selection of the 95% level of confidence is by convention.
Traditionally,
many statistical analyses allow for a 5% error rate. If the exercise of building
the 95% upper confidence limit were to be repeated many times for the facility in
question, 95% of the time this upper limit would exceed the true unknown overall
average level of particle concentration.
Other less commonly employed levels of
confidence are 90% and 99%. Intermediate and more extreme levels of confidence
are possible as well. Selection of an extremely high level of,confidence greatly
increases the risk of failing the Standard, even when the average at all
locations sampled meets the nominal class limit or U descriptor.
The tradeoff
involved with this second type of risk is the reason for not simply choosing an
arbitrarily high confidence level.
The rationale behind the second rule is to require greater uniformity in results
(less variability from location to location). It also reduces the chance of
falsely passing the Standard when limited data are required by the Standard.
When trouble is encountered with this rule, there are two possible statistical
fixes for the problem.
These include (a) collecting data at more locations and
(b) taking additional measurements at one or more locations.
This second statistical rule affects only small facilities, those for which fewer
than ten locations need to be sampled. For such facilities, this rule makes
passing the Standard more difficult than it would be otherwise, especially when
only two locations are required. The thrust of the second rule is to have a high
degree of confidence that the grand average of the cleanliness of the air in the
entire facility is less than the stated.class limit or U descriptor.
39
FED-STD-209E
September 11, 1992
—
From a statistical standpoint, the Standard is not targeted at defects that
impact small areas relative to the sampling plan (e.g., leaks in filters).
If
there were many such leaks, it is likely that some would be detected; however,
the likelihood of detecting any single leak is small.
E30. Sequential samplinq.
For situations in which exactly 20 particles would be
expected in a single measurement taken at the class limit or U descriptor, this
Standard allows the option of using a sequential sampling plan (see 5.1.3.4.4 and
Appendix F). Sequential sampling may reduce on average the sample volume (and
At most, the truncated
therefore the time) required for making each measurement.
sequential sampling plan described in Appendix F will result in the collection of
the full sample volume that would otherwise be obtained if a single sampling plan
were in effect.
.:
If the air being sampled is much cleaner or much more contaminated than the class
limit or U descriptor, the sequential sampling plan will require (on average)
dramatically smaller sample volumes per measurement.
~ven when the cleanliness
of the air is at or near the specified limit, some economy is normally achieved
by using the sequential sampling plan. Furthermore, for a given measurement, the
sequential sampling plan typically provides (prior to testing), a nearly
identical probability of that measurement’s either exceeding or failing to exceed
the class limit or U descriptor as compared tout’he fixed sample size approach.
The main limitations of the sequential sampling plan are: (a) the plan applies
only when the Standard is targeted at exactly 20 particles per measurement at the’
class limit or U descriptor, (b) each measurement requires additional monitoring
and data analysis (although this can be minimized through computerization), and
(c) the average particle concentration as calculated from a given measurement
typically will not be determined as precisely (a direct result of collecting a
smaller sample volume).
0’
Result (c) has an impact on the Standard’s statistical rules. On average, it
will. be somewhat more difficult to pass the upper 95% confidence limit rule (for
fewer than 10 locations) when the sequential sampling plan is used.
E40. Sample calculation to determine statistical validitv of a verification.
The data and calculations presented in the following paragraphs are intended to
serve as a working example, illustrating the statistical procedures involved in
the verification of air in cleanrooms and clean zones. The example is based upon
air sampled for particles 0.3 pm and larger in an effort to verify that the air
sampled complies with”airborne particulate cleanliness Class M 2.5 (Class 10),
for which it is required (Table I) that the UCL not exceed 1060 particles, 0.3 #m.
and larger, per cubic meter.
.
The data for the example, presented in the table in E40.1, include the measured
particle concentrations, ~, obtained for different numbers of samples, N, taken
at each of five locations, L. The calculated average particle concentrations, ~,
at each location are also listed in the table. Calculations follow in E40.2
through E40.5 of: the mean of the average concentrations, M, the standard
deviation of the averages, SD, the standard error of the mean of the averages,
SE, and the upper control limit, UCL.
e
40
FED-sTD-209E
September 11, 1992
E40.I
Tabulation of data.
[email protected]~ at
eachloedon
ticle ConcentnUion9
c,
Laeation 1
1
2
3
4
5
2
530
1200
w
1400
0
I
NR
8s0
100
640
950
3
NR
320
420
320
210
I
4
I
NR
530
850
1200
0
5
I
NR
NR
NR
210
NR
Average
coneentntion
ateachloeadon
N
4
I
1
4
4
5
4
530
725
503
754
290
(NR: no reading taken)
E40.2
Mean of the averaaes.
‘M = (~+~+
M=
E40.3
... +AL)/L
(Equation 5-2)
(530 + 725 + !303+ 754 + 290)/5 = 560
Standard deviation of the averaues.
SD=
(~-M)a+(~-M)a+. ..+(AL-M)a
L-1
(Equation 5-3)
(S30-560)2+ (725-560)2+(503-560)2+(754-560)2+(290-560)2
5-1
SD = 188
E40.4
.
Standard error of the mean of the averaues.
(Equation 5-4)
SE_ 188.84
G
E40.5
*
Urmer confidence limit (UCL~.
UCL = M + (UCL Factor x SE)
(Equation 5-5)
For 5 locations, the UCL factor is 2.13 (see Table II).
*
UCL = 560 + (2.13
X
84) = 739
E40 .6 Conclusion.
Since the 95% upper confidence limit (UCL) is less than 1060
and since the average particle concentration, ~, at each location is lees than
1060, the air sampled is verified as complying with airborne particulate
cleanliness Class M 2.5 (Class 10), even though some of the individual particle
concentrations are above 1060.
41
FED-sTD-209E
September 11, 1992
APPENDIX F
SEQUENTIAL SAMPLING: AN OPTIONAL METHOD FOR VERIFYING THE COMPLIANCE OF AIR
TO THE LIMITS OF AIRBORNE PARTICULATE CLEANLINESS CLASSES M 2.5 AND CLEANER
F1O. “-.
This appendix presents a sequential sampling plan which may be used
to verify the cleanliness of air to airborne particulate cleanliness Classes
M 2.5 and cleaner (Classes 10 and cleaner). This plan matches the properties of
this Standard’s single sampling plan (see 5.1.3.4.1 through 5.1.3.4.3), which
requires a sample duration sufficient to produce an expected 20 counts (E = 20)
in air with a particle concentration exactly at the class limit or U descriptor.
Use of sequential sampling may reduce substantially the sample volumes required
at each location.
F20.
References.
F20.1 Cooper, D. W., and Milholland, D. C., “Sequential Sampling for
Federal Standard 209 for Cleanrooms,” J. Inst. Environ. Sci., U(5)~
28-32
(1990) .
F20.2 Duncan, A. J., Qual itv Control and Industrial Statistics, 4th cd., Irwin,
Homewood, Illinois (1974).
F20.3 Siegmund, D., “Estimation Following Sequential Tests,” Biometrika, B,
341-349 (1978).
F20.4
Wald, A., Sequential Ana~vsis, John Wiley & Sons, New York (1947). .
F30. Backcrround. The advantages of sequential sampling in comparison to single
sampling have been described and demonstrated (see F20.4).
In sequential
sampling, the running total of the particles counted is compared with a reference
count limit that is a function of the amount of sampling done. Sequential
sampling typically requires less sampling than ”any single sampling plan having
This particular
the same probability of false acceptance and false rejection.
sequential sampling scheme has been presented in the literature for use in this
Standard (see F20.1).
F40. Method.
Figure F.1 (see F20.1) illustrates the boundaries of the
The
sequential sampling plan that has been designed- for use in this Standard.
observed number of counts, C, is plotted vs. the expected number of counts, E,
A full single
for air which is precisely at the class limit or U descriptor.
sample corresponds to E = 20. The domains in Figure F.1 were derived from
published formulas (see F20.4 and F20.2). The upper and lower lines are:
Upper:
C =
3.96 + 1.03 E
Equation F40-1
Lower:
C = -3.96 + 1.03 E
Equation F40-2
42
FED-STD-209E
September 11, 1992
25-
I
I
I
20 –
REJECT
I
I
I
*
EXPECTED
COUNT
F. 1.
Observed count, C, vs. expected count, -B, for
sequential sampling. The labels delineate those regions where
the cumulative observed count indicates that the air either exceeds
the class limit or U descriptor (REJECT), meets the class limit or
U descriptor (ACCEPT), or is indeterminate (CONTINUE COUNTING).
Figure
Table F.1 gives the upper and lower times at which C = 0, 1, 2, etc., mean PASS
or FAIL, derived from these equations. . The times are listed both in terms of
E(E=
20 for full sample) and t (t = 1.00 for full sample corresponding to
B = 20).
This plan has been truncated by design so that the conventional single-sample
time (E = 20) is its longest time. A third limit was found to be needed,
c = 20, to match the operating characteristics for the single-sample plan.
8
,
e
To use the sequential sampling plan, record the number of particles observed as a
function of time. Compare the count, as sampling continues, with the upper and
lower limits, using either equations F40-1 and F40-2, or a chart such as Figure
F.1, or Table F.1. Computerized analysis of the data is usually helpful.
If the cumulative observed count for the sample crosses one of the upper lines,
then sampling is stopped and the air is judged to have FAILED. If the cumulative
observed count crosses the lower line, then sampling is stopped and the air is
judged to have PASSED.
If the cumulative observed counts equal 20 or less at the
end of the sample duration, not having crossed the upper line, the air is also
judged to have PASSED.
43
FED-sTD-209E
September 11,” 1992
●
Table F.1. Upper and,lower limits for time at which C counts
should arrive. Times are given in units of expected count
(E = 20 at the class limit or U descriptor) and as the fraction
of total time (t = 1 at the class limit or U descriptor).
FAIL IF COUNT, C, COMES
EARLIER THAN EXPECTED
Count, C
o
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Expected, E
-------------
0.038
1.010
1.981
2.951
3.922
4.893
5.864
6.834
7.805
8.776
9.747
10.718
11.689
12.660
13.631
14.601
15.572
FAIL
Time, t
-----------------
0.0019
0.0505
0.0992
0.1476
0.1961
0.2447
0.2932
0.3417
0.3902
0.4388
0.4873
0.5359
0.5844
0.6330
0.6815
0.7300
0.7786
1.0000
PASS IF COUNT, C, COMES
LATER THAN EXPECTED
Expected, E
3.844
4.815
5.786
6.757
7.728
8.699
9.669
10.640
11.611
12.582
13.553
14.524
15.495
16.466
17.436
18.407
19.378
PASS
PASS
PASS
PASS
PASS
;
Time, t
0.1922
0.2407
0.2893
0.3378
0.3864
0.4349
0.4834
0.5320
0.5805
0.6291
0.6676
0.7262
0.7747
0.8233
0.8718
0.9203
0.9689
1.0000
1.0000
1.0000
1.0000
1.0000
,.
e
An equivalent method is to compare the time at which count C occurs with the
times shown in Table F.1. If the count occurs earlier than expected, as
If the count occurs later
indicated in Table F.1, then the location FAILS.
than expected, as indicated in Table F.1, then the location PASSES. This
requires at most 21 comparisons of arrival time of particles with the limiting
times.
Where a first sample leads to a decision to PASS or FAIL, but it is desired to
test the air further, subsequent samples should be taken as single samples of
duration such that E = 20 for each. Then combine the counts of the single
sample(s) at the location with those of the sequential sample and divide the sum
The result is the
by the total volume of the sequential and single sample(s).
mean concentration at that location, to be compared to the concentration for the
class limit or U descriptor.
44
,
FED-STD-209E
September 11, 1992
To determine the concentration for the entire clean zone from a set of sequential
samples at the designated locations, divide the total number of particles counted
by the total volume of air sampled. A more advanced method of treating the data
can be found in the literature (see F20.3).
Fso.
.
*
ExamDle9.
For further investigation, a reference in the literature
(see F20.1) gives three examples of procedures for sequential sampling.
F60. Rewrtinq.
The data should include an identifier for each location, the
volume of air sampled and the count for each sample, and whether the location
passed or failed for that sample. If a location that failed is resampled, then
the total counts and the total volume (of both the original sample and the
repeated sample) should be reported along with the concentration derived from
their ratio.
45
FED-STD-209E
September 11, 1992
APPENDIX G
SOURCES OF SUPPLEMENTAL
INFORMATION
G1O . sg9E2!2* This appendix lists organizations and other sources from which
supplemental information may be obtained for instruction or guidance in preparing
documents related to the design, acquisition, testing, operation, and maintenance
of cleanrooms and clean zones.
G20.
Sources of SUPPlemental
,-.
.4
information.
G20.1
American Institute of Aeronautics
370 L’Enfant Promenade, S. W.
Washington, DC 20024
G20.2
American National Standards Institute (ANSI)
11 West 42nd Street, 13th Floor
New York, NY 10036
G20.3
American Society of Heating, Refrigerating,
Engineers (ASHRAE)
1791 Tullie” Circle, NE
Atlanta, GA 30329
G20.4
American Society of Mechanical Engineers
345 E. 47th Street
New York, NY 10017
G20.5
American Society for Testing and Materials
1916 Race Street
Philadelphiar PA 19103-1187 ~
G20.6
Association pour la Prevention et l’Etudie de la
Contamination (ASPEC)
Secretariat
d’ASPEC
1, Cite Paradis, ,rue Paradis
75010 Paris
France
G20.7
British Standards Institution
2 Park Street
London WIA ‘2BS
England
G20.8
Commission of the European Communities (CEC)
Office for Official Publications of the European Communities
2 rue Mercier
L 2985, Luxembourg
G20.9
Defence Research Establishment,
National Defence
Ralston, .Alberta
Canada
46
and Astronautics
(AIAA)
and Air-Conditioning
(ASME)
(ASTM)
(BSI)
Suffield
(DRES)
-,.,
m
FED-STD-209E
September 11, 1992
G2O.1O
Deutsches Institut fur Normung (DIN)
(German Institute for Standardization)
Postfach 1107
Burggrafenstrasse 6
1000 Berlin 30
Germany
G20. 11
Food and Drug Administration (FDA)
Division of Drug Quality Compliance
Center for Drugs and Biologics
5600 Fishers Lane
Rockville, MD 20857
G20.12
Institute of Electrical and Electronics Engineers
445 Hoes Lane
PO Box 1331
Piscataway, NJ 0885S-1331
G20.13”
Institute of Environmental
940 E. Northwest Highway
Mount Prospect, IL 60056
Sciences (IES)
G20.14
International Organization
1, rue de Vamsmbe
Case,Postale 56
CH-1211 Geneva 20
Switzerland
for Standardization
G20.15
International Society of Pharmaceutical Engineers
3816 West Linebaugh Avenue
Suite 412
Tampa, FL 33624
Japan Air Cleaning Association
Tomoe-Ya Building No. 2-14
l-Chome, Uchi-Kanda
Chiyodaku, Tokyo, 101
Japan
(JACA)
G20.17
Japanese Standards Association
1-24-4, Akasaka
Minato-Ku
Tokyo, 107
Japan
(JSA)
G20. 18
National Technical Information Service (NTIS)
U. S. Department of Commerce
5285 Port Royal Road
Springfield, VA 22161
G20.16
r
*
-
47
(IEEE)
(1S0)
(ISPE)
y.
...
FED-STD-209E
September 11, 1992
G20.19
Nordic Association for Contamination
R3-kansliet, Paronvagen 15
s-262 62 Angelholm
Sweden
G20.20
Schweizerische Gesellschaft
Seestrasse 5
Postfach
CH-8700 Kusnacht ZH
Switzerland
G20.21
Society of Automotive Engineers
400 Commonwealth Drive
Warrendale, PA 15096
G20.22
Standards Association of Australia
Standards House
80 Arthur Street
North Sydney, NSW 2060
Australia
G20.23
Verein Deutscher Ingenieure (VDI)
VDI-Gesellschaft Technische Gebaude’Ausrustung
Graf-Recke Strasse 84
4000 Dusseldorf 1
Germany
Control
(R3-”NORDIC)
fur Reinraumtechnik
(SAE)
(SAA
.?[
48
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement