ASTM E3336-22

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
Designation: E3336 − 22
Standard Test Method for
Physical Integrity Testing of Single-Use Systems
This standard is issued under the fixed designation E3336; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope 1.8 This standard does not include methods to determine the
maximum allowable leakage limit for maintaining the barrier
1.1 The test methods described in this standard are appli-
properties of the SUS. For that, refer to Practice E3244 and
cable for single-use manufacturing equipment, further called
Test Method E3251.
Single-use Systems (SUSs), used for (bio)pharmaceutical
products. 1.9 This standard does not describe how to select the
appropriate test method. For that, refer to Practice E3244.
1.2 The test methods described in this standard are not
intended to be used on single-use technology for primary 1.10 Furthermore, it does not discuss whether an integrity
containers, combination products (products composed of any test should be conducted, at what frequency and where in the
combination of a drug, device, or biological product), or life cycle of a SUS. For that refer to Practice E3244.
devices. Appropriate procedures related to these products are
1.11 Filter membrane integrity testing that additionally tests
discussed in documents covering the integrity assurance for
the integrity of the SUS is excluded from the scope. Certain
primary containers (1) or medical products (2-4).
components of the SUS may require additional testing.
1.3 The test methods and their validation are described to
1.12 This standard does not purport to address all of the
only cover testing of empty and dry SUSs. Residual liquid in
safety concerns, if any, associated with its use. It is the
the SUS can impact the test reliability and reproducibility.
responsibility of the user of this standard to establish appro-
priate safety, health, and environmental practices and deter-
1.4 The test methods are intended to be used to confirm the
mine the applicability of regulatory limitations prior to use.
barrier properties of the test article, further called integrity
testing, or test the SUS for leaks of certain sizes, further called
1.13 This international standard was developed in accor-
leak testing.
dance with internationally recognized principles on standard-
NOTE 1—To verify that an integrity test can confirm the intended barrier ization established in the Decision on Principles for the
properties of the SUS, its detection limit must be equal or better than the
Development of International Standards, Guides and Recom-
respective maximum allowable leakage limit.
mendations issued by the World Trade Organization Technical
Barriers to Trade (TBT) Committee.
1.5 The physical test methods covered by this standard are:
1.5.1 Pressure-based test methods.
2. Referenced Documents
1.5.2 Tracer gas-based test methods.
2.1 ASTM Standards:
1.6 The physical test methods described are in general
E3244 Practice for Integrity Assurance and Testing of
non-destructive and allow further use of the SUS.
Single-Use Systems
NOTE 2—Some variations can be used in a destructive way, for
E3251 Test Method for Microbial Ingress Testing on Single-
example, to perform root cause analysis of the leak.
Use Systems
1.7 The standard describes the test apparatuses, operation
F2095 Test Methods for Pressure Decay Leak Test for
procedures, environment requirements, and discusses specific
Flexible Packages With and Without Restraining Plates
challenges with testing SUSs, as well as how to perform robust
F2338 Test Method for Nondestructive Detection of Leaks
validation of the test method.
in Packages by Vacuum Decay Method
F2391 Test Method for Measuring Package and Seal Integ-
rity Using Helium as the Tracer Gas
This test method is under the jurisdiction of ASTM Committee E55 on
Manufacture of Pharmaceutical and Biopharmaceutical Products and is the direct
responsibility of Subcommittee E55.07 on Single Use Systems.
Current edition approved Feb. 1, 2022. Published May 2022. DOI: 10.1520/ For referenced ASTM standards, visit the ASTM website, www.astm.org, or
E3336-22. contact ASTM Customer Service at service@astm.org. For Annual Book of ASTM
The boldface numbers in parentheses refer to the list of references at the end of Standards volume information, refer to the standard’s Document Summary page on
this standard. the ASTM website.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
E3336 − 22
2.2 Other Documents: 3.1.9 integrity assurance, n—a holistic approach of risk
USP <1207> Sterile Product Packaging Integrity Evaluation, analysis and mitigation by means of product and process
United States Pharmacopeia (USP), 2016 robustness, quality, and process control and integrity testing to
EU GMP, Annex 1 Manufacture of Sterile Products, Euro- assure that a SUS maintains its integrity prior to and during
pean Commission, 2009 use.
EU GMP, Annex 2 Manufacture of Biological Medicinal
3.1.10 integrity test, n—a test used to confirm the defined
Products for Human Use, European Commission, 2018
barrier properties of a SUS.
USP <1> Injections and Implanted Drug Products (Parenter-
3.1.11 leak, n—a breach in a SUS’s material or a gap
als) – Product Quality Tests, United States Pharmocopeia
between SUS’s components through which there is a break-
(USP), 2020
down of the barrier property of interest.
3.1.12 leak test, n—a test used to identify leaks not corre-
3. Terminology
lated to the defined barrier properties of a SUS.
3.1 Definitions:
3.1.13 maximum allowable leakage limit, n—the greatest
3.1.1 apparatus, n—a technical equipment or machinery
leakage rate (or leak size) tolerable for a given product package
needed for leak or integrity testing purposes.
to maintain its barrier properties under its use-case conditions
3.1.2 balanced pressure drop, n—the pressure drop of posi-
(for example, prevent any risk to product safety, product
tive control articles balanced against the pressure drop of
quality, or operator and environmental safety).
negative control articles; for analyzing the validation results,
3.1.13.1 Discussion—In this test method’s context, the
once this value turns into positive it means that a defective test
product package is a SUS containing a (bio)pharmaceutical
article can be differentiated from a non-defective one.
product, but not a final dosage form.
3.1.2.1 Discussion—The pressure drop values are mean
3.1.14 negative controls, n—the negative control articles are
values of all test articles used for the validation, and include,
intact, under intended use-case conditions non-leaking SUS
depending on the quality requirements of the test validation, a
(see also positive control).
defined number of standard deviations. In addition, the accu-
3.1.15 non-destructive test method, n—a test method that
racy of the measurement instrument should be taken into
maintains the test article in a condition for further use, without
consideration.
impacting its quality attributes (see also destructive test
3.1.3 bioprocess container (biocontainer), n—a container
method).
(bag, bottle, tank, etc.) used primarily for liquid (or frozen
3.1.16 positive controls, n—the positive control articles are
liquid) storage during various stages of biopharmaceutical
test articles of the exact same design as the negative control
manufacturing processing.
articles, equipped with a calibrated leak of known size (see also
3.1.4 calibrated leak, n—a hole which is characterized by its
negative control).
size (for example, artificially created into a SUS, a SUS’s
3.1.16.1 Discussion—Positive controls can be manufactured
material, or component and used for creating positive controls).
by including a calibrated leak into the SUS or by attaching it
3.1.4.1 Discussion—Often, the size is a nominal size which
using an appropriate connection.
is equivalent to a gas flow through an idealized geometry (1).
3.1.17 single-use components, n—parts used in single-use
A commonly used idealized geometry is the “nominal diameter
systems, most commonly, but not limited to, bioprocess
orifice size”, corresponding to the size of a perfect circular hole
containers, tubing, connectors, clamps, valves, sensors, and
of negligible length that would give the same gas flow in the
filters.
calibration conditions (for example, dry air flow rate measured
3.1.18 single-use system (SUS), n—process equipment used
at 25 °C, with 1 bar inlet pressure and 1 atm outlet pressure).
g
in (bio)pharmaceutical manufacturing, disposed of after use
3.1.5 destructive test method, n—a test method that will
and usually constructed of polymer-based materials.
alter the intended use of the test article during the test and not
3.1.19 SUS supplier, n—a manufacturer that produces
allow further use (see also non-destructive test method).
and/or assembles single-use systems, also known as a system
3.1.6 end user, n—a company processing (bio)pharmaceuti-
integrator.
cal products.
3.1.20 tracer gas, n—a gas to be detected against the
3.1.7 family approach, n—an approach to validate only one
background of all other gases.
set of test parameters for a combination of several test article
3.1.21 tracer gas calibrated leak standard, n—element
designs.
emitting a known flow of tracer gas, used to calibrate tracer gas
3.1.8 hardware support structure, n—a hardware that me-
leak detectors. It is an assembly of a pressurized reservoir with
chanically supports the SUS.
an isolation valve and an orifice.
3.1.8.1 Discussion—This can be a restraining hardware, for
3.2 Abbreviations:
example, a pair of plates or grids of rigid material, for example,
3.2.1 BPOG—Biophorum
aluminum or stainless steel, that are used to restrict the
3.2.2 BPSA—Bio Process Systems Alliance
inflation of the SUS, or a hardware that does not restrict the
inflation of the SUS to its nominal volume. 3.2.3 cGMP—current Good Manufacturing Practice
E3336 − 22
3.2.4 CQA—critical quality attributes more detailed visualization of test setups and tested SUS
designs, refer to Practice E3244 and more illustrative technical
3.2.5 HLD—helium leak detector
guides (5).
3.2.6 ICH—International Conference on Harmonization of
Technical Requirements for Registration of Pharmaceuticals
5. Pressure Based Test Methods
for Human Use
5.1 Test Methods Principles:
3.2.7 LoD—limit of detection
5.1.1 The basic principle of a pressure test is to detect leaks
3.2.8 MALL—maximum allowable leakage limit
in the SUS by applying a defined pressure with air (or
sometimes a specified gas).
3.2.9 QbD—quality by design
5.1.2 The flow of gas through any leaks in the SUS can be
3.2.10 QRM—quality risk management
detected either by a pressure decay method after isolation of
3.2.11 SUS—single-use system
the supply pressure or by direct flow measurement at a constant
3.2.12 SUSI(T)—single-use system integrity (testing) system pressure using suitable equipment upstream of the test
article.
3.2.13 SUT—single-use technologies
5.1.3 Both pressure decay and flow measurement tests
3.2.14 TGD—tracer gas detector
depend on the ideal gas law (see Note 4) PV=nRT.
NOTE 4—PV=nRT with P: Pressure, V: Volume, n: Number of moles, R:
4. Significance and Use
Ideal gas constant, T: Absolute temperature.
4.1 The test methods outlined in this standard allow for
5.1.4 Higher pressure, lower test volume and longer test
suppliers and end users of SUSs in (bio)pharmaceutical manu-
time enhance sensitivity, while constant temperature and mini-
facturing processes to detect a leak and/or confirm the barrier
mal ambient air convection is required during the test.
properties of empty, clean, and dry SUSs. Performing integrity
However, pressure ratings of the test article should not be
testing can be a significant contribution to the overall integrity
exceeded. Viscoelasticity and gas permeation characteristics
assurance of SUSs.
should be considered.
4.2 The two types of physical test methods outlined in this
NOTE 5—By using hardware support structure to mechanically support
standard are:
the SUS during the test, it might be possible to apply a test pressure above
4.2.1 Section 5, Pressure-Based Test Methods.
the pressure rating. Furthermore, specific components with limited pres-
4.2.2 Section 6, Tracer Gas-Based Test Methods.
sure resistance (for example, membranes of single-use pressure sensors)
must be protected against over pressurization.
NOTE 3—Other test methods are currently being adapted for robust,
NOTE 6—In order to enhance the test method sensitivity, for complex
reliable, and reproducible testing SUS, for example, Vacuum Decay Test
SUS designs it might be useful to separate the test article into several parts
Method as described in Test Method F2338.
(for example, by clamping the tubing) and test these parts individually.
This could also allow for testing individual parts with a higher pressure
4.3 Pressure-based test methods are generally less sensitive
than other ones.
compared to tracer gas-based test methods but have a lower
complexity and cost. To assist in selecting a method that will fit
5.1.5 In the following sections, statements related to the
an application, refer to Table 1 in Practice E3244 for a more term pressure decay are synonymous for flow rate.
detailed comparison of the two methods.
5.2 Apparatus:
4.4 Both types of test methods can be used to detect leaks of
5.2.1 A measuring instrument that provides the following:
any sizes in a SUS (referred to as leak testing) or confirm the
5.2.1.1 A sensor to detect pressure changes with sufficient
barrier properties of the SUS (referred to as integrity testing).
sensitivity to detect theoretical leak rates according to the
specification of the leaks to be detected in the SUS.
4.5 To ensure that integrity testing performed on SUSs is
5.2.1.2 A timer to control pressurization of the SUS to a
effective and accurate, the properties of the SUS (pressure
pre-set pressure, stabilize the pressure for a set time, and
capabilities, volume, material properties, etc.) must be consid-
provide a time period during which pressure change is re-
ered. Also, a validation should be performed on the chosen test
corded.
method as further described in 5.11 and 6.11.
5.2.1.3 A means to set pressure.
4.6 Practice E3244 should be referenced to determine the
5.2.1.4 A means of holding and displaying the pressure
maximum allowable leakage limit for a SUS, along with the
change inside the SUS between begin and end of the test cycle.
routine testing requirements that are suitable for each applica-
5.2.1.5 A means (optional) to set pressure decay limits for a
tion.
test recipe and alert the operator if the limit is exceeded.
4.7 The purpose of the described test methods is not to
5.2.1.6 A means to connect the test article in a leak tight
stress the SUS until a potential defect occurs. The testing
manner, so that an inflation pressure can be applied to the SUS
parameters, mainly test pressure, are independent from the
and changes in internal pressure can be sensed.
use-case conditions. The robustness of the SUS under use-case
conditions should be proven during product qualification. NOTE 7—It is important to verify the tightness of the entire testing
device, so that it does not contribute to the pressure changes sensed during
4.8 This standard test method describes the test method
testing. For example, this can be done by doing one test without the test
principles, the apparatus designs, and method validations. For article connected.
E3336 − 22
5.2.1.7 A means (optional) to detect and avoid overpressur- 5.5 Preparation of Apparatus:
ization potentially caused by a malfunction of the apparatus, 5.5.1 The apparatus should be placed in a temperature stable
that could lead to rupture of the SUS.
environment (for example, not close to air conditioning or
5.2.2 A hardware support structure (optional) that mechani- direct sunlight) to avoid any drift in the pressure reading during
cally supports the test article when fully inflated or restricts its
the test. Limits for environmental influencing factors should be
inflation that provides the following: assessed during validation of the test method as described in
5.2.2.1 Sufficient pressure resistance to not damage the
5.7.2.6.
support structure when the pre-set test pressure is applied on 5.5.2 The measuring instrument should warm-up after
the SUS.
switch-on to reach a constant apparatus temperature during
5.2.2.2 A structured surface in contact with the test article operation.
allowing gas escape through a potential leak to avoid masking
5.5.3 Connect the measuring device to compressed gas
a leak in the supported surface area of the test article.
supply. Gas supply must be oil-free, dry, and free of particu-
lates. This gas supply must be sufficient to maintain adequate
NOTE 8—Alternatively, a porous layer can be used between the surface
and stable test pressure.
of the SUS and the hardware support structure to allow sufficient gas
escape through the potential leak. The effectiveness of this structured
NOTE 11—Some measuring devices may have a built-in air compressor
surface or porous layer should be confirmed during the leak test
as gas supply. However, requirements on the test gas remain the same.
validation. When selecting appropriate material as a porous layer, the
mesh width and the wire diameter should be considered as critical
5.5.4 Surfaces of a potential hardware support structure
parameters that have an impact on the probability of detection. Re-use of
should be clean and dry.
the porous layer as well as the impact on further processing steps (for
5.5.5 Apparatus should be checked to be leak tight accord-
example, heat transfer) should be evaluated.
ing to 5.2.1.6.
NOTE 9—The hardware support structure can be adjustable (optional) to
optimize the volume of the test article.
5.6 Validation of Test Method:
5.3 Challenges and Potential Interference:
5.6.1 For non-destructive testing, the absence of impact of
5.3.1 As pressure is a function of temperature, environmen-
the test on the CQA of the SUS must be validated.
tal conditions, especially temperature fluctuations, can have a
5.6.2 The test method must be validated as a limit test using
significant impact on the pressure drop measurement.
positive and negative control articles following result interpre-
Therefore, the apparatus and the test article should not be
tation as described in 5.11.
placed in areas facilitating immediate temperature changes, for
5.6.3 Following elements are key parameters to be covered
example, close to a window subjected to direct sunlight or
by the method validation:
close to an air conditioner.
5.6.3.1 the test pressure,
5.3.2 A drift in pressure drop reading can also occur due to
5.6.3.2 the stabilization time,
temperature changes of the test gas. Therefore, it is recom-
5.6.3.3 the test time.
mended to use test gas at the same temperature as the testing
5.6.4 To use the method as a limit test, following consider-
environment.
ations should be evaluated during method validation:
5.3.3 As pressure is a function of volume, flexibility of
NOTE 12—Most of these considerations are derived from ICH Q2 (R1)
polymeric material can result in a volume change of the test
and USP <1207.1> requirements. Additional ones are derived from usual
article during the test and therefore impact the pressure drop
statistical analysis of a binomial distribution. For quantitative tests, please
measurement. It is recommended to choose a stabilization time
refer to ICH Q2 (R1) and USP <1207.1> requirements.
sufficient to compensate for these expansion processes.
5.6.4.1 Accuracy: it corresponds to the percentage of test
5.4 Sampling, Test Articles, and Test Units:
articles correctly classified.
5.4.1 The sample quantity for a method validation is chosen
5.6.4.2 Recall: sometimes called “sensitivity” of the test. It
to permit an adequate determination of representative perfor-
corresponds to the number of positive controls correctly
mance. Positive and negative controls should be used to define
classified divided by the total number of positive controls;
the acceptance criteria.
when the test method is used as an integrity test, it is very
5.4.2 The sample quantity for routine testing (for example,
important to set the acceptance limit at a level maximizing the
statistical or 100 % testing) should be based on a QRM
recall, for obvious reasons, even if it impacts the precision.
approach, as described in Practice E3244.
5.6.4.3 Precision: it corresponds to the number of test
5.4.3 Unique sample identification should be made prior to
articles correctly classified in the population of rejected test
testing to allow the operator to refer to specific test articles, if
articles (classified as fail).
necessary. Information such as test results and anomalies
5.6.4.4 Repeatability: is verified by applying the test method
should be traceable to individual articles.
on multiple sampling of the same homogeneous sample
5.4.4 The identical SUS design should be used for method
population, using the same testing conditions (same operator,
validation as to be used in the routine testing.
limited period of time, same instrument); it is important to
perform this test on a sufficient quantity of samples.
NOTE 10—A family approach is possible with testing the extremes of
each family. Discussion on how to define appropriate families is provided
5.6.4.5 Ruggedness: is verified, if relevant, by having dif-
in Appendix X1.
ferent operators performing the test, having the test performed
5.4.5 The identical apparatus and setup should be used for at different days and using different instruments to perform the
method validation as to be used for routine testing. test.
E3336 − 22
choice of acceptance criteria is provided in 5.11.
5.6.4.6 Specificity: corresponds to the ability of the method
to provide adequate differentiation between true negative and
5.7.2.6 Recording of environmental conditions (for
true positive test articles, despite potential interfering factors.
example, temperature, atmospheric pressure variation) during
Examples of potential interfering factors are described in 5.3.
method validation should be done and appropriate limits
5.6.4.7 The validation should be performed by testing a mix
should be derived as pre-requisite for routine testing.
of positive and negative samples in random order.
5.8 Reagents and Materials:
5.6.4.8 Reproducibility, comparing the output of different
5.8.1 Compressed air – supply cylinder and regulator.
labs, is generally out of scope of such validation.
5.8.2 Nitrogen (nominally 100 %) Gas – supply cylinder
5.6.5 The validation report should ideally include all the
and regulator.
above elements, plus a detailed description of how the positive
5.8.3 For SUS that will be further sterilized, and that are
samples were made, how the defects present in the positive
subjected to a non-destructive routine testing (test on SUS
controls were calibrated or certified, and what are the elements
before use in biopharma manufacturing), following elements
considered to define the most severe cases of the validated
must be considered:
space.
5.6.6 System suitability (also known as performance verifi-
NOTE 17—This situation corresponds typically to a test performed by
the SUS manufacturer, at the end of the assembly step.
cation test) can be performed at the beginning and end of each
testing sequence for added method assurance. Especially,
5.8.3.1 Grade of the gas supply: equivalent to high purity
probabilistic test methods may require a routine demonstration
(>99.99 %) or medical grade, with certificate of conformity;
that the operator is able to successfully differentiate test articles
remaining gases are other gases that naturally can be found in
without defect from those with leaks (ranging in size from
air, at an equal or lower concentration.
smallest to largest, located at various positions), in a blinded
5.8.3.2 Filtration of the gas supply, to address the risk of
challenge study.
particulate matter.
5.7 Calibration:
NOTE 18—Particulates could also cause malfunction of the apparatus,
5.7.1 The measuring instrument should be calibrated and
for example, a built-in calibrated leak for calibration could be blocked.
qualified for its use.
5.8.4 For SUS that are sterile, and that are subjected to a
5.7.2 To achieve a robust and reliable test method, and to
non-destructive routine testing, following elements must be
eliminate as many interference as possible, the method valida-
considered:
tion should be done considering the following steps.
NOTE 19—This situation corresponds typically to a test performed by
5.7.2.1 To eliminate material and configurations-specific
the end-user, before using the SUS in the drug manufacturing process.
interference (for example, creeping of the material) negative
controls, meaning integral SUS, should be tested to establish a 5.8.4.1 Gas supply must be compliant to regulatory
baseline reading for non-defective SUS. requirements, for its production and its monitoring. This
5.7.2.2 To define the method sensitivity positive controls, includes performing periodically a microbial monitoring of the
meaning intentionally compromised SUS with a calibrated leak gas at point of use.
of a known size, should be tested to establish the reading for 5.8.4.2 Filtration of the gas supply, with a sterilizing gas
defective SUS.
filter, to maintain the sterility of the SUS and address the risk
of particulate matter.
NOTE 13—Calibrated leaks could, for example, be laser-drilled holes,
glass capillaries, or microtubes. Leak sizes should be stated normalized in
5.9 Conditioning:
nominal diameter orifice size or leak rate.
5.9.1 Test article should be conditioned to obtain the same
NOTE 14—To confirm the effectiveness of the structured surface or
temperature conditions as exist for the test apparatus. Since
porous layer, if used, the calibrated leak must be placed at the worst-case
measured pressure change is also a function of temperature,
location for potential leak masking.
then the test articles must be at a stable temperature. Testing
5.7.2.3 A family approach can be used to validate a broad
should be done under typical SUS manufacturing environment
range of different configurations. In this case, the extremes of
conditions. All conditions should be recorded at the time of the
each family should be used as positive and negative controls to
test.
validate the parameter set.
NOTE 20—As seen in the combined gas laws, the pressure change is a
NOTE 15—Rationales to define the extremes of families is discussed in
function of temperature. Test articles and the test gas should be at similar
Appendix X1.
temperatures.
5.7.2.4 Depending on the required level of assurance, sta-
5.10 Procedure:
tistical calculation with mean values and standard deviations
5.10.1 Test Article Preparation—As residues in the SUS
(for example, mean 6 3σ) should be applied on the test results,
could block a potential leak, the SUS should be tested empty,
for positive and negative controls respectively.
clean, and dry. In case of testing a pre-sterilized SUS, appro-
5.7.2.5 The acceptance criteria for test evaluation should be
priate measures (for example, pressurizing the SUS through a
placed in between these two set of curves and not interfere with
sterilizing grade filter) must be taken to maintain the sterility.
the lowest value of the statistics for positive controls and
NOTE 21—To maximize sensitivity of the test, the smallest internal
highest value for statistics of the negative controls.
volume of the SUS is desired. See Note 9 about the optional hardware
NOTE 16—More detailed information on statistical calculation and support structure.
E3336 − 22
5.10.2 Apparatus Preparation—The apparatus should be (2) A significant pressure drop can be improved by increas-
prepared as described in 5.5. Appropriate test parameters ing the stabilization time or it indicates a problem in the test
should be defined according to 5.8.1.
setup.
5.10.3 Select and set the test pressure.
(3) A significant negative pressure drop (pressure increase)
5.10.4 Select and set pressurization, stabilization, and test
indicates a problem with environmental testing conditions or
time.
malfunctioning testing device.
5.10.5 Select and set pressure decay limits (if available).
(4) The standard deviation can be much larger than the
5.10.6 Place the SUS in its hardware support structure (if
mean value, as the mean values should be close to zero.
available).
(5) As defined in 5.7.2.4, the standard deviation multiplied
5.10.7 Connect the test article to the measuring instrument
to achieve the target confidence interval is added to the mean
in a leak-tight manner.
value (for example, for x¯ + 3σ to achieve a final
neg neg
NOTE 22—Depending on the complexity of the insertion into the confidence interval of 6σ).
hardware support structure and the connection to the measuring
5.11.1.2 Positive Controls:
instrument, 5.10.6 and 5.10.7 can be done vice versa.
(1) Mean values should show a steep and continuous slope
5.10.8 Begin the test by activating the timer controls and
with a significant pressure drop at the end of the test time.
valves to inflate, stabilize, and measure the test pressure inside
(2) An insignificant pressure drop can be improved by
the SUS.
increasing the test time or test pressure. It can also indicate the
5.10.9 Observe the pressure decay at the end of the test time
test method’s incapability to detect the selected leak size.
period and note if the pressure decay limit has been exceeded.
(3) Along the time-based recording the standard deviation
NOTE 23—Choice of times depends on test article variables and leak
should only be a fraction of the mean values, as the mean
rate requirements. For example, small changes in initial test pressure may
values should increase constantly.
occur from flexible package stretch, thus slightly increasing its volume
(4) As defined in 5.7.2.4, the standard deviation multiplied
(decreasing its pressure) or from fixture contact or the expanding gas
to achieve the target confidence interval is subtracted from the
medium. Increased stabilization time will allow these effects to become
stable before the test data period begins. Test times are selected based on
mean value (for example, for x¯ – 3σ to achieve a final
pos pos
required leakage rates or pressure decay criteria along with the SUS
confidence interval of 6σ).
volume. See 5.11.1 for detailed explanation how to setup appropriate test
5.11.1.3 Determining the acceptance limit by calculating the
parameters.
balanced pressure drop:
5.11 Calculation or Interpretation of Results:
(1) To determine the acceptance limit, time-based statisti-
5.11.1 For the validation of the test method mean values and
cal calculations as described in 5.11.1.1(5) and 5.11.1.2(4)
standard deviations of the time-based pressure drop recording
have to be compared to calculate the balanced pressure drop.
should be calculated and interpreted in the following way to
(2) As all measurements for non-defective test articles as
find a suitable parameter set for a reliable and repeatable
well as for the defective ones are subjected to the accuracy of
differentiation of defective from non-defective test articles.
the measurement instrument, twice the accuracy should be
5.11.1.1 Negative Controls:
taken into consideration as the minimum gap between statistics
(1) Mean values should not show a steep slope and
significant pressure drop at the end of the test time. If the of positive and negative controls for calculating the balanced
stabilization time is chosen appropriately pressure drop should pressure drop.
remain close to zero. (3) As shown in Fig. 1, the minimum test time required to
FIG. 1 Example of Balanced Pressured Drop With Selection of Minimum Test Time
E3336 − 22
reliably differentiate positive from negative controls is the 6.1.3 Several leak detection methods are possible using
point where the balanced pressure drop turns and remains TGD:
positive (for example, 0 < (x¯ – 3σ ) – (x¯ + 3σ ) –
pos pos neg neg 6.1.3.1 Spray test method – local detection (out-in): the test
2*accuracy ).
article is connected to the TGD and slowly challenged at the
(4) As shown in Fig. 2, according to the calculation
target points by a spray of tracer gas delivered from the outside
provided in 5.11.1.3(3) an acceptance limit for the maximum
with a spray pistol.
allowed pressure drop should be chosen at that point of test
6.1.3.2 Sniffer test method – local detection (in-out): the test
time that shows a clear differentiation between negative and
article is pressurized with the tracer gas and controlled from the
positive controls.
outside at the target points by a test gas probe connected to the
5.11.2 For routine testing the obtained pressure drop at the
TGD.
end of the test time has to be compared to the acceptance
6.1.3.3 Chamber gas enrichment test method – global de-
criteria defined according to 5.11.1.
tection (out-in): the test article is connected to the TGD and
5.11.2.1 A pressure drop exceeding the acceptance criteria
surrounded by an enclosure (flexible or rigid) that is filled with
indicates a failed test and a SUS with a leak at least the size as
the tracer gas.
chosen during the test method validation.
5.11.2.2 A pressure drop below the acceptance criteria
6.1.3.4 Chamber accumulation test method – global detec-
indicates a passed test and a SUS with no leak or a leak of
tion (in-out): the test article, placed in a test chamber, is
smaller size as chosen during the test method validation.
pressurized with the tracer gas; the tracer gas leaving the test
5.11.2.3 Unexpectedly high pressure drops can indicate an
article accumulates in the test chamber and is measured with a
improper or leaking test setup. Make sure the test setup is leak
sniffer probe after a defined time period.
tight according to procedure described in 5.2.
6.1.3.5 Chamber vacuum test method – global detection
(in-out): the test article, placed in a test chamber under
6. Tracer-Gas Based Test Methods
vacuum, is pressurized with the tracer gas; the tracer gas
6.1 Test Methods Principles:
leaving the test article is measured by the TGD connected to
6.1.1 Leaks are detected by measuring the presence of a
the test chamber.
tracer gas entering or exiting the test article. The tracer gas
6.1.3.6 Bombing test method – global detection (out-in): the
detector (TGD) is measuring the quantity of tracer gas
test article is placed in a pressure vessel, filled with the tracer
molecules, with calibration this is converted into a volumetric
gas; it is left several hours in the pressure vessel, to let the test
flow rate of the tracer gas.
gas enter in the test article through leaks; the test article is then
6.1.2 Current standard provides detailed steps for test meth-
placed in a vacuum chamber connected to a TGD and the tracer
ods using helium as tracer gas. Other tracer gases can be used,
gas flow escaping from the test article is measured, similarly to
providing adequate assessments are made to successfully cover
6.1.3.5.
the differences with helium.
6.1.4 Amongst the above test methods, only the chamber
test methods (6.1.3.3, 6.1.3.4, 6.1.3.5) and the bombing test
method (6.1.3.6) can be qualified as deterministic, quantitative
Accuracy of the measurement device as mentioned in 5.11.1.3(2).
FIG. 2 Example of Acceptance Criteria Selection (1.0 mbar at 800 s test time)
E3336 − 22
methods. The spray test method and the sniffer test method are the test article) and the test chamber, before injecting the
probabilistic, quantitative methods, useful to localize the area helium and after the measurement.
of a leak.
NOTE 25—In order to achieve a sensitivity sufficient to cover microbial
6.1.5 Current standard provides detailed steps for the sniffer
integrity, deep vacuum evacuation is required. When operating under such
(6.1.3.2) and the chamber vacuum (6.1.3.5) methods. Detailed
a vacuum, the dilution of helium in the remaining air can typically be
neglected.
steps for the other methods can be easily derived from these
two cases.
6.2.2.4 Pressurized helium gas supply.
6.1.6 The sniffer test method is typically used during
6.2.2.5 Pressure probe(s) to measure the absolute pressure
investigations, in development, validation or routine use
in the test article and in the vacuum chamber.
phases, to identify the location of leaks.
6.2.2.6 Connection between test article and helium gas
6.1.7 The spray and the gas enrichment test methods are
supply.
well adapted to rigid parts but can present challenges with
6.2.2.7 Calibrated leak.
flexible ones; they require the test article to be put under
NOTE 26—Such calibrated leaks are used in development, validation,
vacuum, and a flexible test article may collapse and isolate
and routine calibration of the apparatus, for example, to create positive
some areas from the TGD. For that reason, they are generally
controls.
not used to test SUSs.
6.2.2.8 Automated system, controlling the different steps of
6.1.8 The chamber accumulation test method requires a
the test and recording the critical parameters and results.
significantly lower investment cost than test methods using a
6.2.2.9 Optional: hardware support structure in the test
vacuum chamber but offers also much lower performances,
chamber.
especially with SUSs, due to the challenge raised by tracer gas
6.2.2.10 Optional: helium dilution system.
permeation.
6.3 Challenges and Potential Interference:
6.1.9 The bombing test method requires a high differential
6.3.1 Helium is an inert gas. It can however represent a
pressure from outside to inside during the first phase of the test,
safety risk when high level of helium generates a significant
and therefore presents the same challenges as spray and gas
decrease in the partial pressure of oxygen in air and makes it
enrichment test methods, plus the challenge associated to
non breathable. This is addressed by having an adequate
permeation. For these reasons, it is generally not used to test
aeration system, potentially supplemented by oxygen monitor-
SUSs.
ing and alarm system.
6.1.10 The chamber vacuum test method is the most sensi-
6.3.2 Helium permeation through the surface of the SUSs
tive test method for SUSs and is also the most expensive to set
leads quickly to levels impacting the leak testing, which can
up and operate. It is a global test method, testing the entire
mask small size leaks. Speed and level of impact depend on the
SUS, not allowing the leak to be localized. In case of
constituent components of the SUS. For example, silicone
investigation, it can be used for localization of leaks by
tubing are very permeable components.
modifying the test article, that is, testing sub-parts of the test
6.3.3 Vacuum chamber test method in non-steady-state
article or masking part of the test article.
mode:
6.2 Apparatus:
6.3.3.1 The vacuum chamber test method offers a possibility
6.2.1 Sniffer method with helium:
to avoid the above interference, by performing the measure-
6.2.1.1 HLD with sniffer probe and calibration system.
ment before the surge of helium flow due to permeation. The
6.2.1.2 Pressurized helium gas supply.
optimum sensitivity is obtained by finding the best trade-off
6.2.1.3 Pressure probe to measure the pressure in the test
maximizing the signal due to the leak, verified with positive
article.
controls, while keeping a low signal due to the helium
6.2.1.4 Connection between test article and helium gas
permeation, verified by negative controls.
supply.
6.3.3.2 This approach requires to perform the helium leak
6.2.1.5 Helium evacuation system (from test article and
measurement in transient state, well before steady state can be
testing environment). The measurement can typically be done
reached, taking sort of a “snapshot” in dynamic conditions. An
under an extractor or safety hood.
automated system with data acquisition is therefore needed to
be able to validate an apparatus adopting such approach.
NOTE 24—Accumulation of helium caused by gas permeation through
polymeric material can rapidly pollute the testing environment and
Specific attention must be given to ensure proper repeatability
therefore impact the test sensitivity and reliability.
of the dynamic conditions of the test, including all parameters
6.2.1.6 Optional: hardware support structure.
associated to the helium injection in the test article.
6.2.1.7 Optional: helium dilution system. 6.3.3.3 Operating in transient state increases the variations
6.2.1.8 Optional: test enclosure that can be flushed with gas
associated to the location of the leak, both inside the chamber
(for example, N ) to decrease the helium background level. and on the SUS itself; that is, the distance in the vacuum
6.2.2 Chamber vacuum test method with helium:
chamber between the leak and the inlet of the HLD and the
6.2.2.1 HLD with calibration system. distance in the SUS between the helium injection point and the
6.2.2.2 Test chamber, designed to operate under full leak have both an impact on the signal measured by the HLD.
vacuum. This variation must be considered during method validation by
6.2.2.3 Vacuum pumps, to evacuate the test article (manda- using positive controls with calibrated leaks placed at different
tory to have full control on the helium concentration injected in locations.
E3336 − 22
6.3.3.4 The helium flow measured by the HLD will be could also cause local dilution of the helium, leading to
influenced by many elements of the apparatus. It is therefore underestimate the leak in that area.
irrelevant to try to calibrate such apparatus with a tracer gas
6.3.8 Some components are not compatible with a TGD test.
calibrated leak standard (see 6.7). Only a correlation for a limit A good example are the aseptic sterile connectors with mem-
test (pass/fail test) can be done, using a calibrated leak placed
branes that will let the tracer gas pass. Other components may
at the most unfavorable location(s) considering the elements have a specification for tightness not aligned with the TGD test,
described in 6.6.3.3(3).
that is, a higher leak level than the acceptance criterium of the
6.3.3.5 A family approach can be used to cover a range of TGD. Such components should be isolated from the rest of the
SUS sizes and configurations, if all elements above are SUS before performing the integrity test.
considered in the family approach.
6.3.9 The size of the SUS that can be tested is limited by the
apparatus used. For the sniffer test method, the size of the
6.3.3.6 Where it could be demonstrated that permeation
does not play a significant role for the range of SUS tested, a support structure dictates the size of the SUS that can be tested.
If no support structure is used, the maximum allowed pressure
more classical approach, operating in steady state mode, can be
in the test article decreases with the size. It is also a function of
followed, including full calibration and quantitative measure-
the shape and materials of construction of the test article. For
ment. This situation has up to now not been encountered for
the vacuum chamber test method, the size of the vacuum
SUS.
chamber obviously limits the size of the test article.
6.3.4 Pressure resistance of flexible components of the SUS
6.3.10 Helium background noise needs to be minimized
is, in some cases, very limited (for example, bag chambers).
during measurement step as it impacts the test sensitivity.
6.3.4.1 Some components can be damaged even by low
Decrease helium background level in the atmosphere (for the
pressure differential between inside and outside the test article.
sniffer test method) or in the vacuum chamber (for the vacuum
This may limit the pressure of helium in the test article, may
chamber test method) can require a long time. Following
require specific controls to prevent too large pressure differ-
elements must be considered to optimize the operability of the
ences during the vacuum and return to atmosphere ph
...